Sun-Induced chlorophyll Fluorescence (SIF) stands out as a promising remote sensing signal for monitoring photosynthesis over time and space. However, its interpretation becomes intricate under stress conditions due to factors such as light absorption and plant adaptations. To derive the quantum yield of fluorescence (Φ_F) at the photosystem from canopy measurements, the escape probability (f_esc) must be considered.

This study compares Φ_F measured at leaf- and canopy-scale to assess the impact of stress responses, using a potato mesocosm heat-drought experiment as a basis. Initially, we evaluated the performance of reflectance-based approaches for estimating red and far-red f_esc through simulations with the radiative transfer model SCOPE. Findings revealed that existing f_esc models inadequately predicted the correct value range for red f_esc especially at canopy level. In this presentation, we will discuss modifications to address this limitation.

Subsequently, modified models for red f_esc and an existing model for far-red f_esc were employed to analyze the dynamics of leaf and canopy red and far-red fluorescence under increasing drought and heat stress. Incorporating f_esc led to a closer agreement between simultaneously measured leaf and canopy SIF signals, with improved r² values for red f_esc (0.3 to 0.49) and far-red f_esc (0.36 to 0.52). Comparing the quantum yield dynamics of red and far-red fluorescence (Φ_F,687 and Φ_F,760) under increasing stress revealed a significant decrease in both leaf and canopy Φ_F,687, as well as leaf Φ_F,760, as drought and heat intensified. Contrarily, Canopy Φ_F,760 did not exhibit the same trend, displaying wider spread and lower median under low stress conditions. We performed a sensitivity analysis of Φ_F,687 and Φ_F,760 to changing leaf-to-sun angle by comparing measurements with and without mesocosm rotation. It revealed a notable increase in the coefficient of variation of Φ_F,760, especially under unstressed conditions. Our findings underscore the necessity for further research to unravel the causes of discrepancies in leaf and canopy-scale Φ_F,760. Conversely, the underutilized Φ_F,687 demonstrates significant potential for evaluating plant responses to drought and heat stress.

How to cite: Wieneke, S., Pacheco-Labrador, J., Mahecha, M. D., Poblador, S., Vicca, S., and Janssens, I. A.: Improving the Interpretation of Sun-Induced Chlorophyll Fluorescence under Stress: Insights from Leaf- and Canopy-Scale Measurements, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-6519, https://doi.org/10.5194/egusphere-egu24-6519, 2024.

Terrestrial ecosystems are undergoing increasingly frequent periods of stress as the climate is changing. Monitoring ecosystem health efficiently requires global spatial coverage and high temporal resolution, which are assets of satellite-based remote sensing. However, conventional optical remote sensing (RS) approaches offer limited potential for the early detection of ecosystem stress, as changes in ecosystem structure and function often need to be substantial in order to be detectable when using reflectance in the visible and near-infrared range of the energy spectrum. Satellite-based RS of sun-induced chlorophyll fluorescence (SIF) offers much greater promise to that end, but there is still no dedicated SIF mission in-orbit, leaving coarser instruments designed for atmospheric measurements, such as Sentinel-5P TROPOMI, as the only option. The use of SIF from such instruments is challenged by the confounding effects of canopy structure and biochemistry. Furthermore, to correctly diagnose whether plants are under stress, SIF needs to be quantified jointly with the energy that is dissipated as heat. This can be potentially done through monitoring changes in reflectance around the green peak, exploited by the so-called photochemical reflectance index (PRI). Another option may lie in constraining the system with information on land surface temperature (LST), but such measurements should be ideally made at the same time as the instantaneous SIF retrievals. Another challenge is that, combining these measurements requires the proper confrontation of very different spatial footprints over potentially heterogeneous landscapes.

This present work reflects some of the results emerging from the AustroSIF project. The overarching goal of AustroSIF is to make present and future satellite-based SIF measurements a sensitive and reliable means for the early detection of ecosystem stress by combining remotely sensed SIF and PRI. In this project, we have gathered time series of ground-based active and passive chlorophyll fluorescence and hyperspectral reflectance from 7 eddy-covariance flux tower sites. At the same time, we collected respective time series of Sentinel-5P TROPOMI SIF data and MODIS MAIAC PRI data, which we complemented with sub-daily LST measurements from MSG SEVIRI. We also collocated datacubes of Sentinel-2 data to quantify the spatio-temporal heterogeneity within the large TROPOMI and MSG footprints. Finally, we also derived a series of senSCOPE simulations enabling us to place all variables in a synthetic environment to test the strength of the SIF-PRI relationship under stress conditions. By combining together the ground measurements, the satellite measurements, and the simulations, we are able to provide a first glimpse of how well the SIF-PRI relationship can be applied in practice over a variety of  ecosystems.

How to cite: Duveiller, G., Hammerle, A., Hänchen, L., Martini, D., Migliavacca, M., Scholz, K., Sakowska, K., Pabon-Moreno, D., Pacheco-Labrador, J., and Wohlfahrt, G.: Exploring the feasibility of early stress detection with sun-induced chlorophyll fluorescence from tower to satellite , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12555, https://doi.org/10.5194/egusphere-egu24-12555, 2024.

Forest ecosystems, spanning approximately one-third of the Earth's landmass, play a crucial role in providing essential ecosystem services. However, their extension and condition are under threat due to the impacts of climate change. While remote sensing holds the potential to assess the condition and functionality of global forests, challenges in methodology and technology hinder the accurate quantitative estimation of forest traits from spaceborne observations. The emergence of new-generation satellites and advanced retrieval techniques offers the prospect of overcoming these obstacles, yet the potential of both the data and models requires further evaluation. In this contribution, we focused on retrieving forest traits from PRISMA hyperspectral spaceborne imagery employing machine learning regression models as well as hybrid approaches. The area we selected for this study is the Ticino Park, a mid-latitude forest located in northern Italy along the Po river. We conducted an intensive field campaign in the park in the summer of 2022 in conjunction with four PRISMA overpasses to collect trait samples for the calibration and validation of the retrieval schemes. The results obtained highlighted the capability of PRISMA images and retrieval models to precisely quantify Leaf Area Index (LAI) (R²=0.91, nRMSE=8.3%), Leaf Water Content (LWC) (R²=0.97, nRMSE=4.7%) and Leaf Mass per Area (LMA) (R²=0.95, nRMSE=5.6%) in forest ecosystems. Less performing but still promising results were obtained for Leaf Chlorophyll Content (LCC) (R²=0.44, nRMSE=18.3%) and Leaf Nitrogen Content (LNC) (R²=0.63, nRMSE=14.2%). The comparison of the trait values in June and early September revealed a significant decline in both leaf biochemistry and LAI, which can be traced back to the stress induced in the Ticino Park by the severe drought that hit Europe in the summer of 2022. This underscores the valuable role of hyperspectral spaceborne imagery and new generation models for monitoring forest conditions.

How to cite: Tagliabue, G., Panigada, C., Savinelli, B., Vignali, L., Gallia, L., Gentili, R., Colombo, R., and Rossini, M.: Forest traits from PRISMA spaceborne imagery, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-18196, https://doi.org/10.5194/egusphere-egu24-18196, 2024.

How to cite: Luintel, N., Ma, W., Ma, Y., Wang, B., Xu, J., Dawadi, B., Mishra, B., and Dorigo, W.: Tracking the dynamics of paddy rice cultivation practice through MODIS time series and PhenoRice algorithm., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-5834, https://doi.org/10.5194/egusphere-egu24-5834, 2024.

In-season information on expected crop yields is important for farmers' crop management and business planning, as well as for the entire agricultural and food sector. In addition, timely information on possible extreme yield losses in specific production regions allows early decisions in European agricultural policy, e.g. on possible aid payments to producers. Yield losses are mainly caused by adverse and extreme weather conditions such as heat, drought, late frost, heavy rainfall and floods as well as by pests and diseases. Such events are difficult to predict and their actual impact on yields depends on a variety of factors. With ongoing climate change, such adverse conditions and the risk of yield losses are likely to increase (Lüttger and Feike, 2018).

Process-based crop simulation models (CSM) simulate crop growth, development and yield formation, taking into account local soil and weather conditions and potential abiotic stress. For local applications, where actual growth conditions and crop management (e.g., sowing date, cultivar, fertilisation) are known, CSM can be utilized during the season to provide insights into expected crop yields. However, for large-scale applications these information are mostly unknown, which hampers site-specific yield predictions on regional or even national scale. Furthermore, the accuracy of site-specific weather and soil data is limited and actual growing conditions may differ from those assumed in a CSM-based assessment based on such data from national databases.

Point-specific current data on actual crop status derived from remote-sensing can be used to fill those data-gaps and inaccuracies (Guo et al., 2018). Utilizing the newly available high-resolution hyperspectral data from the Environmental Mapping and Analysis Program (EnMAP), a radiation-transfer-model is used to derive pixel-specific state variables of LAI. The actual LAI values are then integrated into a CSM for in-season crop yield predictions on pixel-level. As remote sensing derived crop status data will only be available in the second half of the project phase, we will first use existing observation data from field experiments to mimic the remote sensing data and establish two common data assimilation routines (ensemble Kalman and particle filter) and respective processing pipelines in an ex-post modeling study. After evaluation of the most promising data assimilation technique, the approach will be extended to develop a German-wide winter wheat yield forecast for the current season by using climate forecast data from the DWD. The approach can later be extended to other crops and crop state variables.

How to cite: Quade, M., Attia, A., Preidl, S., Baatz, R., Borrmann, P., and Feike, T.: Remote sensing data assimilation for in-season wheat yield predictions, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15267, https://doi.org/10.5194/egusphere-egu24-15267, 2024.

Solar-induced fluorescence (SIF) is an important indicator of terrestrial photosynthesis and an increasingly targeted observable in spaceborne remote sensing. Here, we present results from efforts to harmonize data across multiple sensors in order to create long-term records that are suitable for multi-decadal analyses. Nevertheless, discontinuities in harmonized records and non-linearities in the relationship between SIF and gross primary production (GPP) demand the use of model enhanced approaches to bridge the gap between observed SIF and inferred GPP. Bayesian model-data fusion (MDF) provides an increasingly established and effective tool to reconcile different satellite datasets and systematically retrieve otherwise unobserved quantities (i.e., not directly measured by spaceborne sensors) such as biomass, leaf area, and GPP, and more accurately estimate interactions between carbon pools and changing climate.

Here, we apply the CARbon DAta–MOdel fraMework (CARDAMOM) MDF system to (i) optimize the parameters and initial states of a terrestrial ecosystem model against global harmonized SIF datasets and ancillary vegetation products to constrain terrestrial photosynthesis and generate reanalyses of GPP, (ii) propagate uncertainty from SIF observations into the GPP reanalyses, and (iii) diagnose these reanalyses to examine the current and evolving state of photosynthesis with climate. Using two SIF products derived from OCO-2 and GOME-2/SCIAMACHY, respectively, we have produced two twenty-year reanalyses of global GPP at a monthly, 0.5-degree resolution. The variability of GPP in these products is well constrained by their respective SIF observations, and reproduces a significant fraction of the observed spatial and interannual variability from globally distributed flux towers. Our results provide a basis for further investigation of photosynthetic carbon uptake as a data product available for the research community.

How to cite: Levine, P., Parazoo, N. C., Bloom, A. A., Yadav, V., Madani, N., Joiner, J., Yoshida, Y., Wen, J., and Sun, Y.: Multi-decadal harmonized records of globally gridded spaceborne fluorescence constrain estimates of terrestrial photosynthetic uptake, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-21437, https://doi.org/10.5194/egusphere-egu24-21437, 2024.

Early and accurate crop yield predictions and prices are crucial for food security management and planning. However, the lack of pre-harvest data poses significant challenges, undermining the reliability and effectiveness of existing methods.
This study introduces an innovative approach that addresses these challenges using satellite data products—specifically, Gross Primary Production (GPP) (0.05° spatial resolution) and dimension-reduction techniques to forecast corn yield and price variation across various regions. We predict national corn yield and price variations by leveraging these satellite-derived products. The value of the approach is demonstrated in three case studies conducted for corn in the US (Corn Belt region), Malawi, and South Africa.
The predictors are derived from GPP year-on-year variation of each region at the peak growing season, i.e., in July for the US Corn Belt (harvest in October) and March for Malawi and South Africa (harvest in May).
We compute the spatial average and Principal Components (PCs) of the GPP year-on-year variations through Empirical Orthogonal Function (EOF) analysis. Additionally, we explore neural network architectures, including Autoencoder (AE) and Variational Autoencoders (VAEs), and extract latent features to reduce the dimension of the GPP data from several thousand to a dozen synthetic features. The PCs, the AE and VAE latent features are used as predictors in Generalized Linear Models (GLM) and Least Absolute Shrinkage and Selection Operator (LASSO) models for predicting year-to-year corn yield and price variation. A neural network is also trained to predict yield and price variations from the latent features for comparison. All models are evaluated using year-to-year cross-validation with three metrics, i.e., Area Under Curve (AUC), the Brier Skill Score (BSS), and the Matthew Correlation Coefficient (MCC).
 Our results demonstrate the superior predictive performance of PCs for US corn yield variations with an AUC of 0.97 (95% CI 0.92-1), a BSS of 0.75, and an MCC of 0.83.
This approach outperforms alternative methods in performance, simplicity, and execution speed. The EOF approach also yields superior results for yield variation prediction in South Africa with an AUC of 0.88 (95% CI 0.75-0.99), a BSS of 0.47, and an MCC of 0.61, while the autoencoder approach is most effective for Malawi with an AUC of 0.98 (95% CI 0.93-1), a BSS of 0.75 and an MCC of 0.83.
For price, our results indicate that the spatial averages of GPP year-on-year July variation in the US Corn Belt can be used to forecast the forthcoming increase or decrease in global corn price at harvest with an AUC of 0.92 (95% CI 0.75-0.99), a BSS of 0.5 and an MCC of 0.66. However, in South Africa and Malawi, the most accurate price predictions are obtained with the VAE approach. With VAE, the AUC is 0.75 (95% CI 0.59-0.92), the BSS is 0.2, and the MCC is 0.27 in South Africa, while these metrics reach 0.94 (95% CI 0.59-0.92), 0.63, and 0.7 in Malawi.
This study highlights the value of combining satellite data with dimension-reduction methods for large-scale prediction of crop yields and price variations several months before harvest.

How to cite: Teste, F., Gangloff, H., Ciais, P., and Makowski, D.: Improving early crop yield and price predictions using satellite imagery with machine and deep learning techniques, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-2028, https://doi.org/10.5194/egusphere-egu24-2028, 2024.

Changes in sub-daily vegetation water content capture the pulse of the Earth's ecosystems. They reflect the interplay between plant function, evaporation, and soil moisture, and underpin land-atmosphere exchange of water and carbon from leaf to global scales. Current and planned microwave missions provide a snapshot every few days. These are adequate to observe inter- and intra-annual variations of above ground biomass (AGB), the slow response in water status over weeks and months, and to map (a-posteriori) biomass loss due to deforestation or mortality. However, they are not sufficient to capture the sub-daily, or even daily, dynamics needed to study ecosystem health.

The SLAINTE (Irish for health) mission aims to fill this critical observation gap at sub-daily scales enabling us to “zoom in” on the fast dynamics associated with water status. Sub-daily observations of VWC are needed to study the vegetation response to the daily cycle in vapour pressure deficit (VPD), the impact of stomatal regulation, and the rate at which vegetation is able to recharge VWC lost during the day. They reveal how ecosystems respond to biotic and abiotic stress (e.g. changing temperature and vapour pressure deficit, soil moisture, insects, disease) and disturbances (e.g. drought, fire). Observing these processes is critical to understand the resilience of terrestrial ecosystems and their water resources in the face of increasing climate variability and extremes, and pressures from human land and water use. The availability of sub-daily SAR data would also fill a critical gap in Earth system knowledge where observations of rapid changes in SSM are essential. For example, they would allow us to observe short-lived wetting/drydown events associated with irrigation, triggering and evolution of flash floods and shallow landslides and the development of hazardous storms.

SLAINTE comprises a small constellation of monostatic L-band Synthetic Aperture Radars (SAR) that will provide sub-daily, ≤1 km scale observations related to ecosystem water status. It has been developed as one of ESA’s New Earth Observation Mission Ideas and was recently submitted in response to ESA’s call for the 12th Earth Explorer. Here, we will provide an overview of the SLAINTE mission idea, our ambitions, and an overview of preliminary science studies. We hope to stimulate discussion with the wider EGU community on how the provision of routine, sub-daily (In)SAR observations could be exploited to address the scientific challenges across the geosciences.

How to cite: Steele-Dunne, S., Bastos, A., Dorigo, W., Massari, C., Milodowski, D., Miralles, D., Ciabatta, L., de Santis, D., Tronquo, E., Zappa, L., Rodriguez Cassola, M., Lhermitte, S., Matar, J., Monteith, A., Taylor, C., Tebaldini, S., Ulander, L., and de Zan, F.: SLAINTE: A sub-daily (In)SAR mission idea to study vegetation water, health and carbon, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17362, https://doi.org/10.5194/egusphere-egu24-17362, 2024.

Vegetation is a key part of the water and carbon cycle and the interaction between Earth's surface and atmosphere. Understanding water dynamics within vegetation is crucial for improving models that represent vegetation processes. Previous studies have investigated exploiting ASCAT scatterometer data from the METOP satellites to evaluate dynamics in vegetation water content. ASCAT has been operational since 2007 and captures microwave backscatter from multiple angles, revealing the relation between backscatter and the incidence angle. This relation reflects the relative contributions of volume and surface scattering—the former affected by water on and within vegetation, and the latter influenced by water in the top soil layer. Currently, a weighted regression using ASCAT observations from 42 days is used to estimate the parameters representing this relation: the slope and curvature, or the first and second order derivative of a second order Taylor approximation, respectively. This estimation method is implemented in the Soil Water Retrieval Retrieval Package developed by TU Wien.  Adverse artefacts of this estimation method are the aggregation of observations corresponding to varying states of the earth surface, e.g. before and after a forest fire. Here, we present results from a study to improve the estimation method for ASCAT's slope and curvature parameters, tailored to quantification of vegetation processes. Goals include: representing parameters at briefer temporal scales, reducing the impact of interception, and restricting temporal aggregation around instantaneous events of change such as storms. In addition to analysing real ASCAT observations, synthetic ASCAT observations are simulated using a radiative transfer model, enabling a thorough comparison of estimated slope against simulated ground truth values. Preliminary results show that simulated ASCAT slope time series represent the dynamics of real ASCAT slope, indicating that synthetic observations can be used to quantify improvement of the slope estimation method.

How to cite: Frantzen, P., Steele-Dunne, S., Vreugdenhil, M., Hahn, S., Quast, R., and Wagner, W.: Improving ASCAT slope estimation methods for representing vegetation water dynamics, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-15193, https://doi.org/10.5194/egusphere-egu24-15193, 2024.

Mountain pastures host rich plant biodiversity organized in various distinct habitats. An accurate long-term monitoring, going beyond the sole ground survey, is of primary importance for nature conservation and forage production planning. We develop here a novel analytical framework to identify and monitor plant communities in mountain pastures based on the joint statistical analysis of ground data and satellite imagery. We use commonly available satellite imagery (Sentinel-2) to track, for the first time to our knowledge, spatial and temporal changes of individual habitats composing the mountain pasture ecosystem, in relation to interannual hydroclimatic variations.

We consider as study zone the mid-to-high elevation mountain pastures surrounding the Swiss National Park in the Grisons canton, Switzerland (approx. 100 sq. km), including fertile pastures, wetlands, dry plant communities, and shrubs. We couple the habitat map to the NDVI spectral index (Sentinel-2 images, ESA) to retrieve a proxy for living vegetation and its productivity. Then, by computing statistical parameters of the NDVI curves, we characterize the annual greening season for the different habitats, taking into account elevational changes and interannual variations of snow persistence, depending on autumn and winter rainfall.

The different habitats show a marked difference in their productivity in function of their wetness until 2400 m a.s.l, while they seem to homogenize at higher altitudes. The greening in the 1-st season half is strongly controlled by snow persistence variations and partially compensated by an after-snowmelt quick growth. Conversely, the 2-nd half season greening is mainly linked to the season maximum NDVI. This workflow presents as an effective strategy to monitor the seasonal and long-term evolution of mountain pasture vegetation in the complex alpine domain.

How to cite: Oriani, F., Aasen, H., and Schneider, M. K.: Monitoring the greening of mountain pasture habitats based on satellite image analysis in response to elevation and seasonal weather change., EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-11269, https://doi.org/10.5194/egusphere-egu24-11269, 2024.

Leaf Area Index (LAI) is a dimensionless measure representing the total leaf area per unit ground area. As such, LAI is a key parameter in crop models, as it directly influences the photosynthetic activity of crops, affecting their growth, development, and yield predictions. It also reflects the canopy structure, which plays an essential role in how the plant responds to environmental stresses. In this study, we used UAV-based light detection and ranging (LiDAR) data and hyperspectral imagery (HSI) as two modalities to predict LAI in a total of 60 plots within 10 wheat fields of various cultivars in the dryland areas of Israel. Field LAI was assessed via two methods – destructive (Li3100C, Licor, Nebraska, USA) and optical (Li2200C, Licor, Nebraska, USA). The LAI in the wheat fields ranged from 0.25 m² m^–2 to 7.7 m² m^–2 (average LAI over the dataset was 1.5 m² m^–2). To predict LAI, we used LiDAR-derived metrics such as height-related metrics (height percentiles and bi-centiles, canopy relief ratio (CRR), max, average, and mode height), and 3-D variables (3-D profile index (3DPI), 3-D voxel index (3DVI), and convex hull volumes), as well as spectral vegetation indices, in five machine learning (ML) algorithms – simple linear regression (SLR), partial least squares regression (PLSR), support vector machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). Single metric SLR resulted in R² ranging from 0.32 to 0.55. More complex algorithms showed that the LiDAR-derived metrics were useful for estimating LAI at the plot level with a higher R² > 0.81 and an RMSE of less than 0.16 m² m^–2 (c. 10%) for the ML algorithms. The 3-D variables were shown to be the most important and robust variables in the ML models for predicting LAI at the plot level, with some height-related variables showing great potential as well. This study is a unique first-step effort to evaluate UAV-LiDAR sensors in collecting high-quality, non-destructive, repeatable measurements of LAI. Such remote sensing information could be highly useful to calibrate and evaluate crop models while resolving the upscaling limitation from leaf to canopy.

How to cite: Mulero, G., Bonfil - Jacques, D., and Helman, D.: Retrieving leaf area index in wheat fields using unmanned aerial vehicle (UAV)-based LiDAR and hyperspectral imagery, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-4589, https://doi.org/10.5194/egusphere-egu24-4589, 2024.

Mediterranean grasslands are essential for the development of rural areas in Mediterranean countries since they provide different ecosystem, social and economic services. Specifically, in Spain, pastures occupy more than 55% of the Spanish surface. The Gross Primary Production (GPP) of this ecosystem is subjected to a natural large spatial and temporal variability due to the influence of the Mediterranean climate.

Remote sensing is accepted as the most powerful tool to study grasslands at different spatial and temporal scales. High frequency satellite data, such as Sentinel-2, offer new possibilities to study grasslands with high spatial (10 m) and temporal resolution (5 days).

Hence, the overall objective of this research is to estimate GPP models for a Mediterranean grassland in central Spain using Sentinel-2 Normalized Difference Vegetation Index (NDVI), complemented with meteorological information at the field scale from January 2018 to August 2020. The GPP models are Light Use Efficiency models and will be validated by the GPP obtained from an eddy-covariance flux tower located in the study site, which belongs to the regional Guadarrama Meteorological Network (GUMNET).

The results shows that the footprint estimation of the flux tower is influenced by mesoscale thermally-driven flows (mountain breezes) due to the presence of the Guadarrama Mountains, located quite close to the station. In addition, pasture phenology is linked to the dynamics of Soil Water Content (SWC), being water the main limiting factor during the growing cycle while temperature is only a limiting factor during winter. Thus, the inclusion of the SWC and minimum temperature in the model provides a better adjustment of the model. With this work we show how the estimated models are adequate to monitor the GPP of this Mediterranean grassland and we present the advantages and limitations found.

How to cite: Cicuéndez, V., Yagüe, C., Inclán, R., Sánchez-Cañete, E. P., Román-Cascón, C., and Sáenz, C.: Modelling Gross Primary Production of a Mediterranean grassland using Sentinel-2 NDVI and meteorological field information , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-12296, https://doi.org/10.5194/egusphere-egu24-12296, 2024.

Flavescence Dorée (FD) is one of the most severe diseases affecting the main viticultural areas of Europe primarily due to a vine-specific leafhopper, Scaphoideus titanus, which indirectly transmits the pathogen to the plant's phloem. In infested grapevine areas where the disease is epidemic and is allowed to spread uncontrolled, epidemic flavescence dorée had catastrophic consequences on yield. Once infected, plants are beyond cure; the only options are severe pruning or total removal. This devastating disease compromises plant growth and can damage entire wine-producing regions, causing substantial economic losses. Delayed symptom recognition contributes significantly to the spread of FD, making early detection strategies essential to effectively manage and mitigate the disease's impact on vineyards.

In this context, remote sensing marks a paradigm shift compared to traditional ground-based surveys. The acquisition of high-resolution images from aircrafts or drones enables efficient scanning of large vineyard areas and detecting subtle changes in leaf color or vigor, enabling faster responses and precise interventions.

In this case study the Radgyro, an experimental aircraft designed for environmental monitoring, surveyed during the initial stages of the disease onset a vineyard of Sangiovese grape variety located in the Emilia-Romagna region (Italy), covering collectively approximately 19 ha with a single 17 min-flight. The centimeter-level resolution of the images acquired by the optical sensors mounted on the Radgyro were automatically processed off-line through a tailored software. The analysis pipeline includes the processing of RGB index maps, where carefully tuned index thresholds were adopted to identify the pixel groups with leaf color attributable to FD symptoms. Spatial clustering algorithms are applied to eliminate noise and isolate potentially diseased plants. The final outputs of the process are the potentially diseased plants, FD density and incidence maps, i.e. prescription maps which provide direct operational guidelines for the FD containment interventions.

On field validation surveys revealed that the process analysis detected 86% of true positive and only 1% of false negative, underscoring an excellent agreement between the remote and ground surveys. Thanks to the high quality of the acquired images and the automatized process analysis, this methodology revealed effective in identifying FD symptoms in the single leaves with a precision comparable to traditional and time-consuming ground-based surveys.

This study was supported by the project PERBACCO (Early warning system per la PrEvenzione della diffusione della flavescenza doRata BAsato sul monitoraggio multiparametriCo airborne delle COlture vinicole) (CUP: E47F23000030002) funded by the Emilia-Romagna Region.

How to cite: Strati, V., Alberi, M., Barbagli, A., Boncompagni, S., Casoli, L., Chiarelli, E., Colla, R., Colonna, T., Franceschi, M., Gallorini, F., Guastaldi, E., Lopane, N., Maino, A., Mazzoli,, G. L., Mantovani, F., Mantovani, F., Migliorini, F., Petrone, D., Pierini, S., Raptis, K. G. C., and Tiso, R.: Airborne surveys for the detection of Flavescence Dorée in vineyards, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10773, https://doi.org/10.5194/egusphere-egu24-10773, 2024.

Mangroves play a crucial role in the global carbon cycle as potential sinks of atmospheric carbon. It has been estimated that the average yearly rate of carbon sequestration by mangrove ecosystems is about two to four times higher than global rates observed in mature tropical forests, thereby making them one of the largest oceanic carbon pools. This importance in the global carbon cycle makes it critical to understand the finer dynamics of mangrove forests starting with detecting and mapping the species of mangrove present in the region.

The mangroves in Senegal are distinctively positioned as one of the few global regions showcasing an upward trend in resilience to climate change, emphasizing their ability to adapt positively to environmental challenges. Numerous studies have confirmed and measured the decade-by-decade growth of mangroves in the Saloum delta through the analysis of remote sensing data. In order to precisely estimate the carbon aggregates, recent research endeavors are focussed on mapping the various mangrove varieties in the region through the examination of multispectral data. This exploration has identified three distinct varieties of mangroves  - Rhizophora racemosa, Rhizophore mangle and Avicennia germinans. The current study seeks to employ Pixxel Hyperspectral data to assess its effectiveness in mangrove zonation and to compare the results with Landsat dataset. Additionally, this study explores the significance of specific wavelength bands in the mapping of mangrove species.

Pixxel is a space technology company building several constellations of the world’s highest resolution hyperspectral earth-imaging satellites. Hyperspectral imagery (HSI) was captured with one of Pixxel’s Technology Demonstration satellites (TD-1) over Saloum delta in Senegal on 09 November 2022, with a spatial resolution of 30 m. The image was processed to Level-2A, surface reflectance data, through Pixxel’s image processing pipeline for atmospheric and geometric correction. A Random Forest algorithm was applied to the surface reflectance data to detect the three species of mangroves present in the delta region. An accuracy of 96.51% was attained with the imagery from TD-1 and 88.55% was achieved using the Landsat-8 dataset having the same spatial resolution for the same region. The identification of the three dominant species of mangroves in the region is consistent with the findings from Lombard et al., 2023. The most significant wavelength bands in distinguishing the different species fell within the Green and  Near-Infrared (NIR) range, with the latter accounting to a larger chunk. The higher classification accuracy from hyperspectral imagery is due to the fact that the large number of narrow spectral bands can distinguish the spectral fingerprints of different species within the mangrove forest. This work has potential to extend beyond classification and extract additional characteristics of mangroves by leveraging the greater spectral information of Hyperspectral Imagery and thus helping us to see the unseen intricacies hidden within the spectra.

How to cite: Sivaraj, P., Yeggina, S., Jalluri, C., and Wright, L.: Mapping Mangrove Zonation in the Saloum Delta in Senegal: Leveraging Pixxel's Hyperspectral Imagery and Assessing Performance against Landsat datasets, EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-17883, https://doi.org/10.5194/egusphere-egu24-17883, 2024.

Accurate, reliable, and up-to-date information on wildlife populations is crucial for biodiversity conservation in the face of unprecedented biodiversity loss worldwide. However, monitoring wildlife populations at large scales remains challenging. Advances in satellite remote sensing, particularly very-high-resolution satellite data, offer new opportunities for monitoring wildlife from space, and new machine learning techniques present great potential for detecting wildlife with remarkable speed and accuracy. Here, we introduce a deep learning pipeline for automatically detecting and counting large migratory ungulate herds (wildebeest and zebra) at the individual level in the Serengeti-Mara ecosystem from submeter-resolution satellite imagery. We apply the pipeline to implement the first-ever population census of large-size ungulates in the Serengeti-Mara ecosystem through a satellite survey and generate the total count of the whole population. The model shows robust performance across diverse landscapes with an overall F1-score of 84.75% (Precision: 87.85%, Recall: 81.86%) on an independent test dataset containing 11,594 animals and achieves good transferability spatially and temporally. This research showcases the capability of satellite remote sensing and deep learning techniques to accurately locate and count very large populations of terrestrial mammals in open landscapes. It provides a new perspective on monitoring wildlife populations and animal migration, which will facilitate the understanding of animal behavior and ecology as well as improve the conservation of the whole ecosystem in the face of rapid environmental changes.

How to cite: Wu, Z., Wang, T., Zhang, C., Duporge, I., Gu, X., Hughey, L., Stabach, J., Skidmore, A., Hopcraft, G., Atkinson, P., McCauley, D., Lamprey, R., Ngene, S., and Gong, P.: Satellite-based monitoring of the world’s largest terrestrial mammal migration using deep learning , EGU General Assembly 2024, Vienna, Austria, 14–19 Apr 2024, EGU24-10882, https://doi.org/10.5194/egusphere-egu24-10882, 2024.