The assessment of species diversity in relatively large areas has always been a challenging task for ecologists, mainly because of the intrinsic difficulty to judge the completeness of species lists and to undertake sufficient and appropriate sampling. Since the variability of remotely sensed signal is expected to be related to landscape diversity, it could be used as a good proxy of diversity at species level. It has been demonstrated that the relation between species and landscape diversity measured from remotely sensed data varies with scale. In this talk, I aim at providing a theoretical and emipircal background of the mostly used diversity indices stemmed from information theory that are commonly applied to quantify landscape diversity from remotely sensed data.

How to cite: Rocchini, D.: Feeling the rhytm of Nature: challenges and prospects of biodiversity prediction from space, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-1449, https://doi.org/10.5194/egusphere-egu24-1449, 2024.

Grasslands and tree-grass ecosystems play a fundamental role in the global carbon balance and the subsistence of the population in vulnerable regions. Protecting the entire range of ecosystem services provided by grasslands require assessing the influence of management and environmental drivers on these services and the role of biodiversity in their provision. Remote sensing offers tools that can help monitoring and better understand biodiversity in grasslands and its relationship with ecosystem function. The spectral diversity hypothesis suggests that spectral variations can be related to functional and phylogenetic diversity. Thus, different authors estimate foliar functional traits and validate the relationship between spectral optical properties of vegetation and functional diversity providing a powerful tool to understand the different roles species play in their environments. These studies mainly focus on forests, while functional characterization of grassland ecosystems is still limited (specially at leaf level) and key leaf traits, such as specific leaf area or cellulose and lignin content, remain underexplored. The phenology of grasslands has been also largely overlooked in biodiversity studies due to the challenges associated to field sampling. As a result, most datasets are collected only over short periods and do not represent the seasonality of the species and associated functional and spectral changes.

In this study, a monoculture experiment was implemented with 7 herbaceous species, including C3 and C4 grasses, legumes and forbs typical of Mediterranean grasslands to assess the capacity of hyperspectral data to detect intra- and inter-specific differences in foliar functional traits of pasture species at different phenological stages, and their plastic responses to water shortage. The experiment included 42 plots (1.5x1.5 m), with six replicates of every other species, organized in two blocks. Water regimes were manipulated to simulate typical versus water stress conditions. Leaf level reflectance was measured using a full range spectroradiometer ASD Fieldspec® 3 coupled with a plant probe and leaf clip with internal light source. Five regular measurements were carried out following the main phenological periods in the spring-summer growing season (April to June) 2022. Besides the reflectance data, key functional traits were also measured including leaf water content (LWC in g/cm²), leaf dry matter content (LDMC in %), specific leaf area (SLA in cm²/g), and chlorophyll and carotenoids concentrations (Cab and Car in mg/g). The potential of optical information to estimate foliar functional traits was explored using empirical models based on Partial Least Squares Regression (PLSR) techniques. Best fits (higher R² and lower normalized root mean squared error (nRMSE)) were achieved for LWC (R² = 0.94, nRMSE = 0.05) and Ca/Cb (R² = 0.89, nRMSE = 0.07), with slightly lower values for SLA (R² 0.71, nRMSE = 0.10). To investigate the seasonal dynamics of functional traits and spectral diversity, hierarchical clustering of the analyzed species based on observed and estimated foliar traits was calculated. Results revealed clear effects of phenology on the spectral diversity and the significant role of the LWC. This variable is not typically used for the functional characterization of herbaceous species.

How to cite: Martín, M. P., Gonzalez-Cascon, R., Burchard-Levine, V., Casillas, L., Rolo, V., and Riaño, D.: Connecting spectral and functional diversity at the leaf-level in Mediterranean herbaceous species: the DiverSpec monoculture experiment, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-16335, https://doi.org/10.5194/egusphere-egu24-16335, 2024.

Grassland ecosystems play a vital role in biodiversity and carbon sequestration, but assessing these ecosystem services accurately is challenging due to their inherent spatial and temporal variability. Conventional field-based methods are often labor-intensive and may not capture this heterogeneity effectively. Recent progress in assessing grassland ecosystems has utilized a combination of structural and spectral data from Unmanned Aerial Vehicles (UAVs), showing promise for a thorough understanding of vegetation behavior. However, this method frequently overlooks an important factor — the horizontal variability within the vegetation, which significantly influences the precision of estimating plant characteristics, particularly in diverse ecosystems. Our study aims to fill this gap by incorporating texture analysis, a critical but often overlooked element in UAV-based assessments. Our research explored the potential of integrating various UAV-derived features to improve the estimation of above-ground biomass (AGB) and species richness in heterogeneous grasslands, key indicators of ecosystem health and productivity. This research investigated the efficacy of combining UAV-derived canopy height, multispectral data, and texture features for AGB and species richness estimation. The study was conducted in a heterogeneous wet grassland ecosystem, using a UAV equipped with multispectral sensors to capture high-resolution imagery. The imagery was processed to extract a range of features, including spectral indices, canopy height models, and textural information using Grey Level Co-occurrence Matrix methods. These features were then used to develop predictive models for AGB and species richness using advanced machine learning techniques, including Random Forest. Model performance was evaluated based on their predictive accuracy and ability to handle the spatial heterogeneity of grassland ecosystems. The study found that models integrating texture analysis with traditional spectral and structural data significantly improved predictive accuracy. For AGB estimation, the best models achieved an R² value of up to 0.84, with a relative root mean square error (rRMSE) of 26.58%. In predicting species richness, the most effective models reached an R² of 0.54 and a relative rRMSE of 31.95%. These results indicate an enhancement in estimation precision compared to models using traditional structural and spectral data types alone. This research demonstrated that UAV-based remote sensing, combined with a fusion of spectral, structural, and textural data, can improve the assessment of grassland characteristics such as AGB and species richness. The findings underscore the potential of integrated UAV-derived datasets in ecological monitoring and highlight the importance of advanced data processing and machine learning techniques in environmental research. This approach offers a promising avenue for more effective grassland management and conservation strategies, contributing to a deeper understanding of ecosystem dynamics.

How to cite: Gonçalves Bazzo, C. O., Kamali, B., Vianna, M., Behrend, D., Hueging, H., Farshid Jahanbakhshi, F., Schleip, I., Mosebach, P., and Gaiser, T.: Innovative Methods in Grassland Monitoring: Integrating UAV Data for Ecosystem Assessment, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15919, https://doi.org/10.5194/egusphere-egu24-15919, 2024.

Vegetation diversity has been found to influence ecosystem function and provide essential ecosystem services, intimately linked to societal wellbeing. However, the relationship between vegetation diversity and function is very complex and still not fully comprehended at different spatial-temporal scales. Indeed, in recent years, remote sensing has shown great promise to better monitor plant diversity at different scales, most commonly through the Spectral Variability Hypothesis (SVH), which links spectral diversity to plant diversity. However, there is still some debate over the generality of the SVH, especially in semi-arid grasslands, which tend to be less studied even though they dominate the trend and inter-annual variability of global water and carbon fluxes. This study focused on examining the relationship between functional diversity (FD) and optical traits of the herbaceous understory of a Mediterranean tree-grass ecosystem (TGE) using field spectroscopy and high resolution imagery from unmanned aerial vehicles (UAVs). Multiple field campaigns were performed from 2021 to 2023 in the Majadas de Tiétar experimental station located in Western Spain to collect in-situ measurements of plant traits (e.g. specific leaf area (SLA), chlorophyll content (Cab)), diversity metrics (functional dispersion (Fdis), Rao’s entropy (Qrao)), hyperspectral field spectroscopy (ASD Fieldspec® 3 portable spectroradiometer) and high-resolution visible-near-infrared (VNIR) and thermal infrared (TIR) imagery onboard UAVs. By applying partial-least-square regression (PLSR) models, high correlations were observed between field spectroscopy and plant traits (r2 > 0.7) with SWIR bands having the most weight in the predictive power of these empirical models, perhaps related to water being the principal limiting factor for herbaceous plants in these semi-arid conditions. By contrast, in-situ PLSR models showed little/no relation to plant diversity metrics (r2 < 0.1). However, preliminary results from the UAV images showed that the spatial heterogeneity of NDVI and land surface temperature (LST), quantified through Qrao using a 5 x 5 pixel window, were positively related to in-situ diversity metrics such as Fdis. Indeed, Qrao based on LST was found to have a more significant relationship to Fdis (p-value < 0.05) compared to Qrao based on NDVI (p-value > 0.05). While the remote sensing of plant functional diversity has concentrated on shortwave reflectances, the use of TIR imagery has large potential as it is more directly related to ecosystem function with its capabilities to act as a proxy for plant transpiration and inform on water use efficiency (WUE). This work is step forward to better understand the optical-diversity relationship in a semi-arid grassland using data acquired at different scales but also from different sources ranging from in-situ hyperspectral measurements to high-resolution TIR imagery. 

How to cite: Burchard-Levine, V., Martín, M. P., Gonzalez-Cascon, R., Rolo, V., Carrascosa, A., Nieto, H., Casillas, L., Riaño, D., and Moreno, G.: Monitoring Grassland Functional Diversity in a Semi-Arid Ecosystem using Multi-Source Close-Range Remote Sensing, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15741, https://doi.org/10.5194/egusphere-egu24-15741, 2024.

Earth observation is a key component for the establishment of biodiversity monitoring systems. The increasing number of instruments acquiring information on Earth surface provides opportunities to assess and monitor various properties of vegetated ecosystems, including vegetation biochemical and biophysical traits and associated processes such as photosynthesis, growth and adaptation to environmental stress. Multiple approaches have been developed during the past decades to link forest taxonomic diversity with remotely sensed information, with varying degrees of success. Statistical metrics directly derived from the vegetation reflectance such as spectral variance have shown limitations for the estimation of taxonomic diversity. One reason is that factors intrinsic and extrinsic to vegetation influence reflectance and contribute to this variance. On the other hand, this reflectance can be converted into optically effective plant properties (optical traits) using statistical methods (e.g. spectral transformation, machine learning or spectral indices) or physical methods (e.g. physical model inversion) applied to optical imagery, with an objective to reduce the influence of extrinsic factors. The spatial heterogeneity of a set of optical traits may then be used as a relevant proxy for vegetation diversity. Statistical methods are computationally efficient, but lack generalization ability, while physical approaches show better potential for generalization ability, but show limitations when applied on complex systems. Moreover, the set of optical traits accessible from optical data varies with sensor characteristics: new imaging spectroscopy missions expand the range of variables for which quantitative assessment is possible compared to multispectral imagery.

We introduce a framework taking advantage of physical modelling to assess a set of vegetation traits then used to feed remotely sensed diversity mapping techniques in the context of forest ecosystems. This approach intends to convert the optical information on a physical basis, in terms of vegetation traits related to structural, compositional and functional properties prior to computing diversity metrics. Physical modeling contributes to minimizing the influence of factors extrinsic to vegetation on optical traits, as a way to improve the generalization ability of existing frameworks taking advantage of Earth observation through space and time. To illustrate it, we used the model PROSAIL to assess Leaf Area Index, leaf chlorophyll content, equivalent water thickness and leaf mass per area from imaging spectroscopy acquired over forested areas. The method implemented in the R package biodivMapR was then applied to compute various diversity metrics from these vegetation biophysical properties, including α- and β-diversity metrics usually obtained from species inventories in ecological applications. We illustrate this framework with data acquired over different sites and with various optical sensors, including airborne and spaceborne imaging spectroscopy, and discuss current limitations.

How to cite: Féret, J.-B.: Mapping forest biodiversity from optical imagery: a plant trait-based method, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-13456, https://doi.org/10.5194/egusphere-egu24-13456, 2024.

The current availability of hyperspectral images (HS) acquired by the PRecursore Iperspettrale della Missione Applicativa (PRISMA) mission of the Italian Space Agency and the recently launched Environmental Mapping and Analysis Program (EnMAP) mission of the German Space Agency, as well as the planned missions, e.g., the Copernicus Hyperspectral Imaging Mission for the Environment (CHIME) of the European Space Agency open unique perspectives for the multi-temporal mapping of forest biodiversity. In this work we use the high spectral resolution of spectral signatures provided by PRISMA images to derive unsupervised maps of vegetation diversity. Study areas are located in the National Parks of Gargano, Alta Murgia, Cilento-Vallo di Diano-Alburni, Appennino Lucano Val D’Agri Lagonegrese and Pollino, all in Southern Italy. Two indexes are used to pre-filter forested areas in HS images: Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI). The high spectral resolution of HS images allows to compute the different combinations of NIR and Red bands, for the computation of NDVI, and of NIR and SWIR bands for the computation of NDWI. This gives a more statistical weight to the thresholding of index maps to identify the areas covered by vegetation. The spectral signature profiles at the pixels, selected based on the index maps, are further processed using the Principal Component Analysis to reduce data dimensionality, and clustered using the K-means algorithm. As a result, a map of the vegetation diversity is obtained, with the location of pixels belonging to the different clusters identified by the K-means algorithms. The set of spectral signatures measured at pixels belonging to the same cluster are used to statistically characterize the reflectivity of vegetation.

ACKNOWLEDGMENTS

Project carried out using ORIGINAL PRISMA Products - © Italian Space Agency (ASI); the Products have been delivered under an ASI License to Use.

How to cite: Nico, G. and Masci, O.: Forest biodiversity mapping based on PRISMA hyperspectral images, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-20068, https://doi.org/10.5194/egusphere-egu24-20068, 2024.

The global biodiversity crisis emphasizes the importance of diversity monitoring to examine ecosystem stability and resilience, including regeneration capacity. As a critical driver of change in tropical mountains, landslides alter the structure, composition, and function of landscapes. One possibility to study the large-scale causes and consequences of landslides on diversity is to use remote sensing to characterize ecosystem traits and functions at several spatial and temporal scales. An increasing availability of satellite-borne hyperspectral offers the possibility to capture morphological and physiological traits of vegetation to characterize functional diversity in areas affected by landslides. Using hyperspectral data to characterize functional diversity often involves the removal of bare soil to eliminate background reflectance. Given that landslides of different ages contain a mixture of vegetated and bare soil pixels, the challenge is to incorporate the latter into image processing, and ultimately into metrics that provide an integrative functional characterization of areas undergoing succession. We define landscape diversity as the structural, functional, and historical characteristics of ecosystems and may be useful to expand functional diversity monitoring beyond purely vegetated areas. Using a historical landslide database and the novel PRISMA (PRecursore IperSpettrale della Missione Applicativa) hyperspectral data we addressed two questions to assess the role of landslide history on landscape diversity. First, do areas with different landslide histories exhibit distinct functional trait and landscape diversity patterns? Second, which landscape traits distinguish landslides of different recovery stage considering landslide size, age, frequency, and reactivation status?

To address these two questions, we derived two sets of variables that represent landslide history and landscape diversity. To represent landslide history, we processed historic and current remotely sensed data from the Sierra de Las Minas (SLM) mountains in eastern Guatemala to create a geodatabase that includes landslide inventories (1973 – 2021) in which each landslide is characterized by age, size, and shape. In ArcGIS Pro we identified degree of overlap among landslides from all inventories to mark landslide reactivation. Specifically, a model identifying landslide overlap distinguished landslides that occur once from landslides that occur repeatedly. To represent landscape diversity, we processed PRISMA data to create functional indices representative of vegetation and soil traits across the SLM. To include areas across all stages of succession, both traits of vegetation and bare soil three separate indices were created. A first masked out bare ground, a second masked out vegetation, and a third version combined the vegetation and bare ground. Finally, we examined the relationship between landslide history and the three versions of functional indices using geographic Random Forest algorithm. The outcomes of this study could reveal lasting structural, compositional, and functional impacts of landslides in tropical mountains, which serve as critical safeholds for biodiversity and ecosystem services during drastic global change.

How to cite: Kilgore, A.: Landslide history shapes landscape diversity: Applying hyperspectral data for functional diversity monitoring of tropical mountainous ecosystems, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17915, https://doi.org/10.5194/egusphere-egu24-17915, 2024.

Remote sensing plays a crucial role in ecology and nature conservation by allowing for effective monitoring of spatio-temporal changes in ecosystems. The NaturaSat software, developed by a multidisciplinary team of botany fieldwork scientists, nature protection managers, mathematicians, and software developers (Mikula et al., 2021), is a recent tool that focuses on vegetation exploration. This software offers solutions to challenging questions, such as accurately estimating the areas and boundaries of Natura 2000 habitats and tracking their spatio-temporal changes using an evolving curves approach. It also enables the monitoring of biodiversity and habitat quality through the use of a graph-Laplacian function. Additionally, the software utilizes a novel deep learning method called the Natural Numerical Network to classify habitats on the most detailed scale represented by vegetation units.

Within this talk, we will present how NaturaSat software has been extensively tested in various habitats, ranging from temperate lowland wetlands along the Danube River to riparian forests, broadleaved deciduous forests, ancient montane woodlands of the Carpathian Mountains, and even the arctic tundra vegetation in the Alaska region. The results have demonstrated the software's capability to accurately identify habitat borders and detect shifts in these borders due to changes in water regimes or climate. Furthermore, it has proven effective in distinguishing between species-rich natural forests and planted forests dominated by the same tree species. The software also enables the classification of habitats and the automatic detection of new habitats in previously undiscovered areas. In conclusion, the NaturaSat software complements detailed ground-based approaches in biodiversity exploration and habitat monitoring. The presented methods can be repeated over long time periods, ensuring temporal consistency, and they offer a cost-effective means of identifying trends in vegetation changes and biodiversity.

How to cite: Šibíková, M., Šibík, J., Šlenker, M., Ožvat, A. A., Kollár, M., and Mikula, K.: Advancing Biodiversity Exploration and Habitat Monitoring: Utilizing the NaturaSat Software for ecosystems from temperate lowland wetlands to Arctic tundra vegetation., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-8583, https://doi.org/10.5194/egusphere-egu24-8583, 2024.

Context :

This work is part of a doctoral thesis being carried out on the marshes of the regional natural park of Brière (France, Loire-Atlantique, 44). The main purpose is to study the changing dynamics of plant formations on the entire sector, in order to enable prospective modeling. The natural park of Brière would like to set up a Biodiversity Observatory in order to gain a better understanding of the functioning of this fast-changing ecosystem. Field sampling efforts are very costly and time-consuming, and are hampered by difficult access to the extensive marsh areas (over 18 000 hectares). As a result, existing maps are not spatially exhaustive and present a simplification of habitats mosaics, making modeling impossible.

Purposes :

The aim of the study is to (1) map the current distribution of plant communities (2) identify their dynamics through the historical evolution of potential habitats in relation with hydrology, invasion by trees, agro-pastoral and traditional practices and (3) deduce the marsh's capacity for stability and resilience in the coming decades.

Materiel and methods :

The first step involves an hyperspectral and LiDAR aerial survey across the whole sector. It is completed by the acquisition of a World View 3 scene, in order to assess the contributions of the different types of data. At the same time, floristic field surveys were carried out to characterize the textural and spectral variability of the images. Automatic classification methods were then applied.
Based on the results obtained from the mapping, we can assess the quality of the biodiversity in the Brière region. Indeed, plant formations differ in terms of stability and resilience.

Main results :

A complete mapping of the 18 000 hectare of marshland has been carried out, according to the Eunis declination, at a spatial resolution of 1.30 meter. This includes areas not yet mapped. In addition to these classic communities, we have also added invasive exotic species, which are very present in the Brière region. The relation between biodiversity loss and human factors from the Middle Ages to the present day has been established.
Standardized data acquisition and processing methods have been set up to enable long-term changes monitoring.

How to cite: Lafitte, T.: Remote sensing for mapping Natura 2000 habitats in the Brière marshes (France, Loire-Atlantique, 44) : setting up a long-term monitoring strategy to understand changes , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17576, https://doi.org/10.5194/egusphere-egu24-17576, 2024.

There exists an urgent need towards forest value chain optimization on societal level. Effective policy making is required to reduce effects of global climate change and or to improve yields in forestry. Unfortunately, these two ecosystem services are competing and even conflicting with each other. To succeed in their joint optimization, precise carbon intake and water balance estimates in different biomes are crucial and require new tools for the task.

Forest data are typically collected with forest inventories. These inventories provide the fundamental information for all decision-making in society and industry that are relevant to human interventions, including harvest planning. Nordic countries have long performed forest inventories on a national level to estimate the country-wide forest averages with area-based inventories (ABA) and at stand level relying on airborne laser scanning (ALS). Current ABA techniques are limited in their spatial resolution and can be improved by focusing on individual tree level. Individual tree level mapping allows to focus not only on the wood material volumes, but also to their quality and health. Computation of this information, especially over wide geographic areas, is a significant computational and data management challenge and requires high-performance computing.

Our goal is to develop the missing mapping technology and demonstrate this in Finland where we will automatically count and characterize all five billion dominant and co-dominant forest trees. We will do this by merging already existing laser scanning technologies on different scales, by developing novel methods as needed, and finally implementing them in the EuroHPC LUMI supercomputer. Each tree will be individually segmented and imputed with wood quality information. All trees and their parameters are collected into the “Metsäkanta” database for interactive mapping and Digital Twinning applications.

We will further enhance the database by computing the CO₂ sink potential for each detected tree. The CO₂ sink potential is modelled from dense spatiotemporal in-situ laser scanning references collected from individual trees. The results are then imputed to all trees. The outcome of these efforts will be a Digital Forest Replicate (DFR) at individual tree level. The DFR combines the information of individual tree wood quality, growth potential, and near real-time carbon sink reporting. This allows improved country-level carbon stock estimates.

How to cite: Puttonen, E., Hyyppä, J., Hyyppä, M., Yu, X., Virtanen, J.-P., Campos, M., Kivimäki, A., and Wang, Y.: Towards nation-wide individual tree carbon sink and biodiversity mapping utilizing high performance computing, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-19757, https://doi.org/10.5194/egusphere-egu24-19757, 2024.