Improving the validation of ecological niche models with remote sensing analysis

Accurate and reliable lithological mapping through satellite-borne remote sensing data and image classification approaches has a critical role since it can automatically and promptly identify lithological units over large areas. Most available Pixel-Object Based comparative classification studies have been applied to land use land cover (LULC) studies; however, this research aims to evaluate and compare the performance of these digital classification methods in the field of geological mapping in semi-arid areas, by integrating spectral bands and neo-bands, particularly the Minimum noise fraction (MNF) and the principal component analysis (PCA), of Sentinel-2A satellite imagery, to map the southern of Skhour Rehamna which is located at the western Moroccan Meseta. The analysis results from two different methods, namely, pixel-based image analysis (PBIA) with k-nearest neighbour (K-NN) and Random Forest (RF) machine learning algorithms (MLAs), and Geographic Object-Based Image Analysis (GEOBIA) were assessed and compared. PBIA method involved selection of training areas whether it was k-NN or RF MLAs, and produced lithological maps that exhibit “salt and pepper” effects as well as problems associated to delineating accurate lithological boundaries, while GEOBIA approach involved multi-resolution segmentation step where scale, shape and compactness parameters should be adjusted as accurate as possible, in order to segment the image into homogeneous and meaningful regions so that the resulted samples were classified using Standard Nearest Neighbour algorithm. Therefore, the resulting lithological maps were assessed by comparing both techniques using confusion matrix, overall accuracy (OA) and Kappa coefficient (K). The results show that the GEOBIA approach had higher overall agreement (83.46% OA and 0.76 K) than RF (81.92% OA and 0.72 K) and k-NN (80.79% OA and 0.70 K) PBIA approaches. Overall, the results clearly indicate the potential of GEOBIA technique for lithological mapping applications to produce more realistic maps.

The mango mealybug Rastrococcus invadens (Williams) (Homoptera: Pseudococcidae) is a destructive and important insect pest of fruit trees in Africa and Asia, especially the mango. Females and nymphs feed on plant leaves and fruits and produce honeydew that causes sooty mold, leading to yield reduction. Although it is an important pest, the distribution of R. invadens under different climate change scenarios has not been established. In this study, we predicted the suitable habitat for R. invadens occurrence under current and future [two Shared Socioeconomic Pathways (SSPs) scenarios: (SSP2-4.5 and SSP5-8.5) for the years 2050s and 2070s], using environmental variables and four ecological niche models viz., maxent, random forest, boosted regression trees, and support vector machines. The performance and accuracy of these models were evaluated using the area under the curve (AUC), the true skill statistic (TSS), correlation (COR), and deviance. All models had high accuracy (AUC ≥ 0.96, TSS ≥ 0.88, COR ≥ 0.74 and deviance ≤ 0.3) in predicting the potential distribution of R. invadens. Among the four models, the random forest algorithm had the highest performance (AUC = 0.99, TSS = 0.95, COR = 0.91 and deviance = 0.14) in predicting the potential distribution of R. invadens, followed by maxent (AUC = 0.97, TSS = 0.90, COR = 0.81 and deviance = 0.22). However, the maxent model was the best among the four algorithms in predicting the ecological niche of R. invadens. The precipitation of the wettest month (bio13) was the most crucial environmental variable that contributed to the predictions of the four models. The results revealed that most areas in East, Central, and West Africa were projected with high suitability for R. invadens to occur under current climatic conditions. Similarly, Bangladesh, Laos, Myanmar, India, Thailand, Vietnam and Cambodia in Asia, as well as Brazil, and Venezuela in South America showed high suitability for R. invadens establishment. However, under future climatic conditions (the years 2050s and 2070s), the suitable areas for R. invadens will increase regardless of the SSPs scenario (SSP2-4.5 and SSP5-8.5) indicating an expansion of the geographical range for this pest. This expansion is projected to be higher for the years 2070s than the 2050s. Similarly, the invasion risk of R. invadens is predicted to be higher under SSP2-4.5 scenario compared to SSP5-8.5 scenario, regardless of the year of the projection. Our results serve as an early warning tool that could serve as a guide to prevent further spread and invasion of this pest to new areas as well as help in developing an effective management strategy against R. invadens.

Populus pruinosa is listed in the International Union for Conservation of Nature (IUCN) Red List of Threatened Species and the Red Data Book of Kazakhstan. The habitats of the species are associated with the floodplains of desert rivers, including the Syr Darya and the Ili. The reduction in tugai forests is associated with changes in the hydrological regimes of rivers, the reduction in groundwater, fires, and grazing. The aim of this research is to advance our knowledge of the woodlands and forests of P. pruinosa in Kazakhstan by analyzing the cenoflora and forestry characteristics, conducting ecological niche modeling to select suitable habitats for ecological restoration, and identifying rehabilitation strategies. The vegetation was studied using traditional methods of geobotanic field research in 2021 and 2022. Ecological niche modeling includes the calculation of all the environmental variables that are significant for the species, taken in their optimal ranges, and identifying the intersection of all the favorable conditions. To identify these variables, geographically linked meeting points of the studied species, climatic and soil conditions, relief characteristics identified by the SRTM and raster sets of the environmental variables (WorldClim), and distances to watercourses and settlements were accepted for consideration. A total of 102 environmental variables were used.

A zona costeira se estende das planícies costeiras até a borda externa das plataformas continentais. A GEOBIA (Geographic Object-Based Image Analysis) é um método eficaz na análise de imagens de alta resolução, incluindo fotointerpretação e classificação de características da paisagem. O método de classificação supervisionada consiste em um conhecimento prévio de algumas áreas a serem estudadas, o que permite amostras de confiáveis. Assim, o presente estudo tem como objetivo principal comparar o método de classificação supervisionada pixel a pixel com o método de GEOBIA, buscando o mapeamento da cobertura da terra no setor Norte da Ilha do Maranhão. O algoritmo classificador utilizado foi o Maximum Likelihood Classification. O método de Classificação Supervisionada apresentou um valor de Kappa de 0,8469, Exatidão Global de 0,8950 e Variância Kappa de 0,0009295. Para a GEOBIA apresentou um valor de Kappa de 0,9410, Exatidão Global de 0,9600 e Variância Kappa de 0,0004075. O sensoriamento remoto se mostra extremamente eficiente em pesquisas relacionadas ao uso e ocupação da terra. Os resultados do método da Classificação Supervisionada, apesar de suas limitações, mostraram resultados satisfatórios quando comparados com os resultados do método da GEOBIA, o que torna interessante sua aplicação para iniciativas de mapeamento em diversos municípios brasileiros.

In this chapter, we describe a procedure for enhancing the automatic image segmentation for land cover mapping from Very High Resolution images. The increased need for large scale mapping (1:10000) for local territorial monitoring led to think about mapping production. Nowadays mapping production for land cover and land use (LUC) is mainly performed with human photo-interpretation. This approach can be extremely time consuming, expensive and tedious for data producers. This is confirmed from the evidence of rural areas where the use of the GIS database for LUC is less numerous than the GIS urban database. In the last decade, Geographic Object-Based Image Analysis (GEOBIA) has been developed by the image processing community. This new paradigm builds on theory, methods and tools for replicating the human photo-interpretation process from remote sensing data (Hay and Castilla 2008). However, the GEOBIA community is still fragile and suffers from a lack of protocols and standards for operational LUC mapping applications. Currently, human photo-interpretation seems a safer option. The objective of this research is to find an alternative to this time consuming and expensive use of human expertise. We explored the limits of GEOBIA to propose an automatic image segmentation enhancement for an operational mapping application. Questions behind this study were: What is a good segmentation? How can we obtain it?

Mapping refined burned areas (BA) in the Brazilian Amazon is still a challenge. The main difficulty of BA detection in large areas is the presence of cloud cover and water bodies. The use of different data sources of medium spatial resolution satellite images can provide a higher availability of cloud-free images. Besides that, it may decrease the uncertainties associated with coarse spatial resolution data (>250m), which can under or overestimate BA and hinder the detection of small BA patches (<0.1km²). In this study, we propose an innovative methodology based on spectral indices and geographic object-based image analysis (GEOBIA), using medium spatial resolution images to improve BA detection in the Brazilian Amazon region. Firstly, we assessed the performance of nine spectral indices in two study areas, derived from Landsat-8 OLI and Sentinel-2A MSI data to identify the most suitable index for BA detection in this region. Then, we refined this data through the GEOBIA-based model. The results showed that the Burned Area Index (BAI) was the most suitable index for BA mapping (M index >1.5) for both sensors. Our model allowed detecting more than 80% of small BA and also presented high Dice coefficient values (~0.70) with low omission and commission errors (0.22 and 0.32, respectively). Such combined approach corresponds to a novel contribution to the BA detection in the Brazilian Amazon region and for enhancing the operational product generation.

AN ASSESSMENT OF SPATIAL VARIATION OF LAND SURFACE CHARACTERISTICS OF MINNA, NIGER STATE NIGERIA FOR SUSTAINABLE URBANIZATION USING GEOSPATIAL TECHNIQUES

Ecological models require a large variety of variables in order to describe the biotic and abiotic conditions that define wildlife habitats and biotope distribution. Instead of providing this information using regular grids or categorical maps, geographic object-based image analysis allows to describe the land cover based on meaningful spatial regions which are closer to the natural habitats than regular grids and more flexible to the specific needs of models than categorical data. In the frame of the Lifewatch/Wallonia-Brussels project, we use GEOBIA to create a geographic object database closely related to the ecological concept of ecotope. Through an iterative process with ecological modellers, this database has been designed in order to define its key characteristics : it had to provide a quantitative description of the land cover within topographically relevant spatial regions and it integrates contextual information from different scales. The proof of concept of this new type of geographic database was done over the Wallon region in Belgium. This study area covers approximately 16800 square kilometers with a very fragmented landscape. A dataset including aerial photographs at 0.25 cm resolution and LIDAR at 0.8 pts/m has been provided by theWalloon region for the study. The data were resampled at 2m resolution for the purpose of the analysis. The data processing workflow includes three steps : pixel-based image classification, image segmentation and object-based integration. Pixel-based image classification consists in a supervised classification with the spectral values from the aerial photographs (NIR/Red/green/blue), the Digital Height Model extracted from the LIDAR and the intensity of the first LIDAR return. This yielded a classification into broadleaved trees, needleleaved trees, grass, bare soil, crop, pavement, building, water and shadows with more than 80% overal accuracy. The image segmentation approach is the main novelty of this research. In order to fit with the biotopes, image segments indeed had to take the type of slope into account. This was achieved by computing pseudo-hillshades for North-South andWest-East orientation and including those two files together with the spectral information from the aerial photographs. The result of this analysis is a set of topographically relevant ecotope delineation. The last step applied contextual decision rules to consistently aggregate the land cover information at the ecotope level and add more information from ancillary datasets.

Protected areas for conservation and poverty alleviation: experiences from Madagascar Charlie J. Gardner*, Martin E. Nicoll, Tsibara Mbohoahy, Kirsten L. L. Oleson, Anitry N. Ratsifandrihamanana, Joelisoa Ratsirarson, Lily-Arison Ren e de Roland, Malika Virah-Sawmy, Bienvenue Zafindrasilivonona and Zoe G. Davies WWF Madagascar and Western Indian Ocean Programme Office, BP738, Antananarivo 101, Madagascar; Durrell Institute of Conservation and Ecology (DICE), School of Anthropology and Conservation, University of Kent, Canterbury, Kent CT2 7NR, UK; D epartement de Biologie, Facult e des Sciences, Universit e de Toliara, Toliara 601, BP 185, Toliara, Madagascar; Blue Ventures Conservation, Level 2 Annex, Omnibus Business Centre, 39-41 North Road, N7 9DP London, UK; ESSA-D epartement Eaux et Forets, Universit e d’Antananarivo, BP 175 Antananarivo, Madagascar; and The Peregrine Fund, BP 4113 Antananarivo, Madagascar

Land use and land cover (LULC) mapping is essential for land-based climate change adaptation and mitigation strategies. This study presents the development of 10-meter high-resolution (HR) land use maps within the RethinkAction H2020 project, aimed at enhancing spatial planning for climate mitigation and adaptation. The methodology integrates multi-source remote sensing data, machine learning classification techniques, and auxiliary datasets to generate accurate and transferable land use classifications across six European bioclimatic regions. The study employs Sentinel-2 and Landsat-8 imagery, using supervised classification with Random Forest (RF) and Geographic Object-Based Image Analysis (GEOBIA) to enhance accuracy and minimize spectral confusion. This approach resulted in the creation of twelve HR land use maps at two classification levels, covering six case study (CS) areas. A key contribution of this research is the generation of suitability maps, which assess the potential for implementing land-based mitigation and adaptation solutions (LAMS) such as reforestation, water harvesting, and photovoltaic energy development. This study highlights the importance of integrating remote sensing, machine learning, and spatial analysis to support evidence-based decision-making in land use planning, offering a scalable and replicable methodology for detailed LULC classification.

This study attempted to predict ground subsidence occurrence using statistical and machine learning models, specifically the evidential belief function (EBF), index of entropy (IoE), support vector machine (SVM), and random forest (RF) models in the Rafsanjan Plain in southern Iran to investigate 11 possible causative factors: slope percent, aspect, topographic wetness index (TWI), plan and profile curvatures, normalized difference vegetation index (NDVI), land use, lithology, distance to river, groundwater drawdown, and elevation. The Boruta algorithm was applied to determine the importance of the possible causative factors. NDVI, groundwater drawdown, land use, and lithology had the strongest relationships with land subsidence. Finally, we generated land subsidence maps using different machine learning and statistical models. The accuracy of these models was assessed using the AUC value and the true skill statistic (TSS) metrics. The SVM model had the highest prediction accuracy (AUC = 0.967, TSS = 0.91), followed by RF (AUC = 0.936, TSS = 0.87), EBF (AUC = 0.907, TSS = 0.83), and IoE (AUC= 0.88, TSS = 0.8).

BackgroundDue to climate change, the geographical distribution of sand flies during the last decades has shifted northward from latitudes below 45°N in southern Europe to latitudes just above 50○N. Recent studies show that some phlebotomine sand flies were recorded in several parts of Germany and Belgium. In central Europe, some autochthone leishmaniasis cases are being recorded in regions traditionally regarded as leishmaniasis-free. An important challenge is to predict the geographical distribution of leishmaniasis vectors under new climatic conditions. In this study, we attempted to predict the current distribution of six leishmaniasis vectors in the Mediterranean basin and forecast species’ geographical shift under future climate scenarios using an ensemble ecological niche modeling approach. Species records were obtained from scientific surveys published in the research literature between 2006 and 2016. A series of climate metrics describing temperature and precipitation in the study area under two climatic scenarios were obtained from WorldClim database. A consensus model was derived from six varieties of modeling approaches (regression, machine learning and classification techniques) in order to ensure valid prediction of distribution of vectors under different climate scenarios.ResultsModel performance was generally high for the included species with a specificity (true negative rate) ranging from 81.03 to 96.52% (mean = 86.94%) and a sensitivity (true positive rate) ranging from 87.93 to 100% (mean = 96.98%). Our work evidenced the hypothesis of the widespread of Leishmania vectors under climate change scenarios. All of the studied species are prospected to gain new areas that are actually not suitable for vectors’ survival. Phlebotomine sand flies are prospected to invade extra-Mediterranean regions, especially western and central Europe.ConclusionsOur study confirmed the importance of environmental and climate factors on the distribution of leishmaniasis vectors and demonstrated the performance of ecological niche modeling in the prediction of the geographical spread of vector-borne diseases. Ecological niche modeling should be considered in the future as a valuable tool in addition to experimental laboratory studies for a better understanding of the biology of vector species.

Niche conservatism providing support for using ecological niche modeling in biological invasions has been widely noticed; nonetheless, the impacts of human disturbance on niche changes and model predictions are not fully recognized. In this study, climate niche changes during the invasion of Japanese beetle (Popillia japonica) were studied under the framework of niche unfilling, stability, and expansion situations. The beetle might undergo niche expansion because of the small native range (Japan) relative to the broad invade range (North America). Relative roles of climatic and anthropogenic factors and their interactions in determining the spread of P. japonica were investigated. Results suggest that the climate niche was conserved during P. japonica invasion, climate niche unfilling rather than niche expansion was observed, suggesting there are suitable climate space remain unoccupied. Inclusion of anthropogenic factors into climate niche model has improved niche model prediction values in areas of high human activity. During P. japonica invasion, anthropogenic factors influenced the climate niche expression and altered the effect of climate suitability. In areas of poor or unsuitable climate, the human activities facilitated the establishment of P. japonica and functioned synergistically, whereas in highly suitable areas, including anthropogenic variables decreased climate suitability. These findings are pivotal to our understanding of climate niche conservatism and niche model transferability when predicting areas of invasion of P. japonica and can be applied elsewhere for predicting other pest’s invasion potential.

International audience

Satellite-derived vegetation indices contribute significantly to the prediction of epiphyllous liverworts