The increasing volume of data at multiple scales—from leaf-level measurements (e.g., gas exchange), tree-level measurements (e.g., sap flow and dendroecology), ecosystem-level measurements (e.g., eddy covariance towers, UAVs, aircraft), to Earth observation from space—presents new opportunities to address these challenges. This session invites studies that improve our overall understanding of biosphere-atmosphere interactions by addressing the mismatches across different temporal and spatial scales and integrating these insights into modeling strategies. We particularly encourage contributions that explore the effects of climate extremes (e.g., drought, heatwaves, excess rainfall, winter warming) on carbon, nitrogen, and water fluxes. In addition to empirical multi-scale observations, we welcome research that delves into data-driven diagnostics and constraints for model evaluation, data-driven parameterisations in mechanistic models, and the development of data-driven/hybrid modelling strategies (i.e., seamless fusion of data-driven approaches and mechanistic models) for an integrated understanding of carbon, nitrogen, and water fluxes across scales.
Orals: Wed, 6 May, 14:00–18:00 | Room 1.31/32
India is the second largest contributor to the global greening, which is situated in the higher carbon uptake tropical region. We find that India has been greening with a marked enhancement in Normalised Difference Vegetation Index (NDVI, 10%), Leaf Area Index (LAI, 11%) and Solar Induced Fluorescence (SIF, 13%) in recent decade (2010 to 2019) as compared to previous decade (2000 to 2009). This greening is largely cropland-based where croplands exhibit twice the greening magnitude of forests. Cropland greening driven by improved irrigation facilities, better farm mechanisation, enhanced land management and use of nitrogen fertilisers contributes 86.5% to India’s net greening. To comprehend the translation of this greening into ecosystem health and functionality, we assess the Carbon Use Efficiency (CUE), and Water Use Efficiency (WUE). Soil moisture (SM) exhibits direct causal relationships with CUE and its determinants, with SM being the primary driver of both CUE and WUE. The coupling of the carbon and water cycles in India has intensified in recent decades, particularly in croplands. We find hindered ability of Indian forests to translate the structure (greenness) into functioning (carbon uptake) recent decades. To further decipher this, we for the first time estimated the Ecosystem Photosynthetic Efficiency (EPE) for Indian forests. Our recent study explicitly highlights the weakening of Indian forests as carbon stocks in the recent decade (2010–2019) from the previous decade (2000–2009) due to reduction in the translation factor i.e., EPE. This decline in EPE is predominant in the pristine forests of eastern Himalaya and Western Ghats due to enhanced moisture stress, rising aridity and increased wildfires, in the warming climate. We find just 16% of the Indian forests maintain high ecological integrity or intactness.
How to cite: Kuttippurath, J. and Kashyap, R.: Changing vegetation-carbon-climate relationships in India during recent decades , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18260, https://doi.org/10.5194/egusphere-egu26-18260, 2026.
To clarify the impacts of drought, high temperature, and nitrogen application on the growth, yield, and grain quality of spring maize, the variety "Danyu 405" was used as the test material. Experiments involving drought stress, high temperature stress, and nitrogen addition were conducted during key growth stages such as jointing, tasseling, and flowering. The study systematically revealed the comprehensive effects of drought, high temperature, and nitrogen fertilization on spring maize (variety 'Danyu 405'). In the water-nitrogen interaction experiments, it was found that compared with the control under adequate water conditions without nitrogen, moderate drought combined with nitrogen application at the jointing or tasseling stages increased plant height, barren tip ratio, amino acid content, and crude protein content. However, it significantly suppressed the leaf area index, biomass, hundred-grain weight, and theoretical yield, while reducing grain fat and starch content. Drought at the tasseling stage was particularly detrimental to yield, with an average reduction of 40.8% in theoretical yield. Furthermore, as nitrogen application increased, most yield-related indicators showed a declining trend, while some quality indicators (such as starch) improved, indicating that nitrogen application under drought conditions could enhance certain quality traits while inhibiting yield. In experiments combining temperature and water stress, it was further demonstrated that drought, high temperature, and their combined stress significantly reduced the maximum carboxylation rate of leaves (by 23.3% to 33.2%) and yield. Drought alone reduced yield by 40.0%, while the combined effect of high temperature and drought was less severe than that of drought alone. Additionally, the combined stress significantly altered grain quality, manifesting as increased fat and amino acid content and decreased starch content. The findings provide a theoretical basis for spring maize production in Northeast China to address climate change and optimize water and nitrogen management.
How to cite: Chen, N.: Effects of Drought, High Temperature, and Nitrogen Application on the Growth, Yield, and Grain Quality of Spring Maize in Northeast China, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15781, https://doi.org/10.5194/egusphere-egu26-15781, 2026.
Forested areas are a primary contributor of the terrestrial carbon sink; however, extreme temperature events, including both warm and cool anomalies, have been shown to influence carbon dioxide (CO2) uptake in forested ecosystems. With the frequency and severity of these extreme temperature events increasing due to climate change, the CO2 uptake response of forested ecosystems to these extreme temperature events may become increasingly impactful. Hemiboreal forests are of particular interest due to their unique geographical location in which the forests’ tree species are situated in their climatic limit; thus, hemiboreal forests have an increased susceptibility to the effects of climate change due to their location in the transition zone between temperate and boreal forests.
To quantify the response of forest-atmosphere CO2 exchange of hemiboreal forest ecosystems to extreme temperature events, we analysed daily eddy-covariance net CO2 fluxes, quantified by the net ecosystem exchange (NEE), measured at Howland Forest (Maine, U.S.A.) and the derived component fluxes gross primary production (GPP) and ecosystem respiration (Reco). Using the 29-year dataset of daily CO2 fluxes and corresponding meteorology, we derived daily CO2 flux and temperature anomalies to assess the impact of extreme temperature events (i.e., days with air temperature greater than 2 standard deviations above/below the mean air temperature for that day of year).
Our results indicate that, on average, net CO2 uptake is reduced in response to both extreme warm and cool events in the hemiboreal forest. We observed a statistically significant decrease in net CO2 uptake (corresponding to positive NEE anomalies) during extreme warm and cool events in seven and five of the nine non-winter months (i.e., March to November), respectively. Particularly in the summer months, both extreme warm and cool events were associated with less CO2 uptake. The NEE response in the non-winter months resulted from reduced GPP during both extreme warm and cool events as well as elevated Reco during extreme warm events. Future analyses will investigate the impact of the frequency and magnitude of extreme temperature events on monthly and annual CO2 budgets of this hemiboreal forest ecosystem.
These results demonstrate a decrease in net CO2 uptake in response to extreme temperature events with potentially negative effects on the CO2 sink strengths of hemiboreal forest ecosystems; this may reinforce a positive feedback loop with increasing air temperature decreasing CO2 uptake in forested ecosystems, contributing again to increasing air temperatures. Our research helps to gain a more thorough understanding into the role that forested ecosystems play in terrestrial CO2 sequestration in today’s changing climate.
How to cite: Seely, P., Ouimette, A., Ruzol, R., and Helbig, M.: Impact of extreme temperature events on CO2 uptake in a hemiboreal forest in North America , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19925, https://doi.org/10.5194/egusphere-egu26-19925, 2026.
The ecological resilience of Mediterranean forests is increasingly challenged by the rising frequency, severity, and unpredictability of drought events driven by climate change (IPCC, 2021). In recent years, flash droughts, characterized by their rapid onset and short duration, have emerged as a significant climatic stressor. Understanding how different tree species respond to these abrupt drought events is essential for predicting forest vulnerability and resilience under future climate scenarios and for translating physiological responses into effective forest conservation and management strategies.
This study aims to: (1) analyze the temporal characteristics of flash drought events, including their onset, duration, and intensity, using atmospheric indicators; and (2) assess species-specific responses of stem radial growth (SRG), tree water deficit (TWD), and stem water transport (sap flow) during and after flash drought episodes, with a comparison between coniferous and broadleaves species.
The research was conducted in the Piegaro Forest, located in the Umbria region of central Italy (42.96°N, 12.06°E; 430 m a.s.l.). The study site is dominated by deciduous broadleaved stands, mainly oaks (Quercus cerris and Quercus petraea) and wild cherry (Prunus avium), alongside conifer plantations of Douglas-fir (Pseudotsuga menziesii) and Scots pine (Pinus sylvestris). An IoT-based monitoring platform, the TreeTalkerCyber device, was installed on selected trees to continuously record individual tree physiological functioning and microclimatic conditions.
Our findings show that flash drought period significantly suppressed stem radial growth in conifer species, whereas this suppression was not significant in broadleaved species, despite that all species maintaining sap flow during the flash drought. Notably, sap flow played a critical role in sustaining growth during flash drought periods. However, the effects of drought stress were more pronounced in the post-drought period, with reduced stem growth and sap flow compared to pre-drought conditions. Overall, stem radial growth emerged as the most sensitive and responsive indicator, revealing persistent internal water stress that extended beyond drought termination.
These results provide valuable insights into species-specific drought resilience and have important implications for sustainable forest management and silvicultural practices under increasing climate variability and the intensification of flash drought events.
How to cite: Samad, N., Mannsfeld, A., Yates, J., and Chiriacò, M. V.: Flash drought alters immediate growth rather than tree water relations: A case study from central Italy., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21125, https://doi.org/10.5194/egusphere-egu26-21125, 2026.
Evergreen needleleaf forests span a wide climatic range, yet their carbon sequestration is increasingly constrained by climate-driven environmental limits. Building on results presented at EGU 2025, which showed a shift in the dominant seasonal control on productivity from atmospheric moisture in the Boreal Hyytiälä forest (Finland, HYY) to soil moisture in the semi-arid Yatir forest (Israel, YAT), we identify the environmental boundary conditions underlying this contrast.
Using PAR-saturated conditions to isolate eco-physiological controls on productivity, we derive key climatic and hydrological thresholds from SHAP-based analyses. In YAT, productivity is strongly constrained by deep soil water availability, with a clear threshold at ~15.8 %vol in the deepest measured soil layer (~45 cm). This threshold reflected a seasonal transition where deep soil moisture shifts from limiting productivity (ineffective water retention and root resistance) to supporting shallow root water uptake and productivity during the wet season. This transition coincides with the seasonal minimum in soil temperature imposing peak root resistance, indicating a compounded control on the onset of productivity in this water-limited ecosystem.
In contrast, seasonal productivity in HYY is dominated by precipitation, which both sustains evapotranspiration, closely linked to net ecosystem productivity (R=0.96), and likely reflects a favorable cloud and radiation regime. The high historical ratio of diffuse to direct shortwave radiation in HYY (Sdiff:S ~ 3:4) helps to buffer canopy conductance against high vapor pressure deficit (VPD), consistent with the high sensitivity observed at this site (negative productivity response at VPD > 1 kPa). Such atmospheric constraints are lacking in YAT, where diffuse radiation is limited (Sdiff:S ~ 1:4) and VPD shows an order of magnitude larger range.
Despite adaptation to such contrasting environments, both forests exhibit a similar optimal air temperature range for productivity (14–20 °C), which highlights a shared physiological optimum across the divergent environmental limitations. Overall, our results demonstrate that carbon sequestration in these systems is not controlled by universal drivers, but by site-specific boundary conditions, such as deep soil water availability in semi-arid Mediterranean forests and precipitation-linked atmospheric regimes in Boreal forests.
How to cite: Rez, L., Vesala, T., Kolari, P., Tziperman, E., Rubin, R., and Yakir, D.: Determining the controlling factors for carbon sequestration in two contrasting forests in the Boreal and semi-arid Mediterranean regions (Part II), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16479, https://doi.org/10.5194/egusphere-egu26-16479, 2026.
Accurately quantifying phenological dynamics in vegetated systems is essential as the timing of seasonal plant activity drives water, nutrient, and carbon cycling, but it is also one of the processes most disrupted by environmental changes. Land Surface Models (LSMs) integrate phenology to represent these feedback loops between terrestrial ecosystem functioning and the global climate, but their performance relies on the type and quality of data used for evaluation. Traditional phenological datasets span from point observations (e.g. leaf-on, leaf-off dates), to high resolution ground measurements of greenness and productivity (e.g. GCC), to time series of remote sensed vegetation indexes (e.g. EVI). However, each observation type measures distinct ecosystem properties, and no single data source provides both the temporal and spatial coverage needed to fully represent phenology at regional and global scales. Here we integrate growing season metrics for 89 European temperate forests sites across scales, derived from eddy covariance measurements, phenocam time series, and MODIS remote sensed vegetation indexes. For the first time, we incorporate phenological dates derived from citizen science, drawing data from the Flora Incognita app, GBIF and iNaturalist, to calculate species-specific annual observation curves for 11 characteristic understory species spanning 2020-2024. By combining data streams, we evaluate the advantages and limitations of each for data-model integration, and further assess the potential of opportunistic species observations to scale up non-overlapping phenological data to ecosystems. By simulating the same sites with the QUINCY LSM we also investigate the role of process-based models to bridge between datasets, as the processes underlying growing season that they provide aid the interpretation of differences between observed phenological metrics. This work highlights the potential of integrating multi-scale phenological information, including underutilised contribution from citizen science, to improve our understanding of phenological dynamics. We additionally explore how LSMs can be leveraged together with data for ecological insight beyond evaluation, moving away from the traditional one-way relationship between data and models.
How to cite: Yajima, M., Daly, L., Rzanny, M., Wäldchen, J., Mäder, P., Nelson, J. A., and Caldararu, S.: Tree and leaf: merging multi-stream data, models and citizen science for phenological detection, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3517, https://doi.org/10.5194/egusphere-egu26-3517, 2026.
Forests worldwide are increasingly threatened by biotic and abiotic stressors, yet many operational monitoring approaches rely on visual surveys or spectral and structural proxies that often reflect anomalies only after physiological dysfunction has become pronounced. Wireless sensor networks such as TreeTalker[1] enable continuous observation of tree physiological and hydrological states that may reveal stress responses before visible symptoms emerge. However, the individual tree-level nature of these observations necessitates upscaling approaches capable of extending physiological insights to unsensed trees across forest stands. In this study, we present thermal infrared remote sensing as a promising method for this purpose, as canopy temperature is theoretically linked to transpiration and stomatal regulation[2].
A physiological and hydrological baseline was first established using TreeTalker observations collected between July 2021 and August 2022 across two temperate forest plantations (Long and Belt plantations, UK). Soil moisture dynamics were characterised using changepoint-based segmentation to describe event-scale drydown behaviour, rainfall response timing, and drought occurrence. At the tree level, soil-stem coupling was quantified using night-time stem water content, while sap flow regulation and recovery were examined using cross-correlation and Granger causality frameworks that incorporated atmospheric demand via vapour pressure deficit. Together, these analyses revealed strong inter-tree and seasonal variability in hydraulic coupling and regulation strategies, providing a baseline characterisation of site and tree condition.
Building on the TreeTalker baseline, thermal surveys were conducted in May 2025 using a laboratory-calibrated DJI Matrice 210-mounted Zenmuse XT sensor over the same plantations. Tree-level mean canopy temperatures were then extracted from processed thermal orthomosaics and compared with contemporaneous TreeTalker measurements of sap flow and stem water content acquired at or near the UAV overpass. Results show that an inverse relationship between canopy temperature and sap flow is more consistently expressed among Belt Plantation trees, although notable exceptions indicate heterogeneous regulation strategies. In Long Plantation, this pattern is not dominant across the population and is largely driven by a single high-sap flow tree exhibiting a markedly cooler canopy, while most other trees show substantial scatter. No clear relationships between canopy temperature and stem water content are observed at both sites. Additional information from visual surveys of tree condition indicates that some trees exhibiting pronounced structural symptoms deviate from the general thermal-sap flow tendency, suggesting that canopy structure may contribute to tree-specific decoupling without explaining site-wide patterns.
These results indicate that UAV-derived canopy temperature primarily reflects instantaneous hydraulic flux and regulatory behaviour rather than buffered internal water storage. By anchoring thermal observations within a multi-season physiological baseline, this work demonstrates how thermal imagery can be used to upscale continuous tree-level sap flow measurements.
References
[1] Valentini, R., Belelli, M.L., Gianelle, D. et al. New tree monitoring systems: from Industry 4.0 to Nature 4.0. Ann. Silvic. Research 43(2), 84–88 (2019). https://doi.org/10.12899/asr-1847
[2] Smigaj, M., Agarwal, A., Bartholomeus, H. et al. Thermal Infrared Remote Sensing of Stress Responses in Forest Environments: a Review of Developments, Challenges, and Opportunities. Curr. For. Rep. 10, 56–76 (2024). https://doi.org/10.1007/s40725-023-00207-z
How to cite: Busari, Q., Gaulton, R., and Brown, P.: Assessing the Potential of UAV Thermal Imagery for Upscaling Tree-Level Physiological Measurements, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15071, https://doi.org/10.5194/egusphere-egu26-15071, 2026.
Forests are pivotal components of the global carbon cycle, yet satellite Earth Observation (EO) products for canopy condition, disturbance, and carbon flux proxies remain challenging to validate in remote and hard-to-reach regions where continuous ground measurements are scarce. The RemoTrees project was launched to bridge this gap by developing and deploying autonomous, low-power IoT, satellite enabled, multi-sensor systems and a harmonized data workflow that couples in-situ observations with EO information to improve the robustness and interpretability of carbon-cycle assessments under increasing climate extremes.
After the initial project phase focused on requirements, baseline demonstrations, and integration concepts, RemoTrees has progressed over the last two years to a technology maturation stage with successive device generations. A beta version of the RemoTrees node has been engineered and validated through laboratory characterization and pilot field deployments, enabling end-to-end testing of sensing, power autonomy, telemetry, remote management, and data continuity in operational forest conditions. These results directly informed iterative improvements leading to a gamma (final) version, targeting higher reliability, easier field maintainability, and improved data quality for EO calibration/validation use cases. Across these iterations, the project has refined a modular sensor approach to capture key variables relevant to forest functioning and stress—combining under-canopy VIS–NIR radiometric observations with complementary eco-physiological and environmental measurements (e.g., soil moisture, sap flow, and microclimate context)—and has strengthened data handling through structured metadata, quality control, and alignment with satellite acquisition constraints.
We present the RemoTrees mid-term status, highlighting the transition from concept to validated beta deployments and the consolidation into the gamma platform. Finally, we outline the next project steps: scaling deployments across different forest types, consolidating calibration and validation protocols, and advancing data-fusion strategies so that continuous ground observations can more effectively reduce uncertainties in EO-based carbon monitoring and support resilient forest management.
How to cite: Coppola, V., Renzi, F., Tagliacarne, F., Yates, J., Belelli Marchesini, L., and Valentini, R.: RemoTrees: advancing scalable ground validation of satellite products with a new generation of autonomous satellite forest sensor nodes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20320, https://doi.org/10.5194/egusphere-egu26-20320, 2026.
Europe has witnessed a significant increase in the number and ferocity of so-called ‘mega-fires’, a phenomenon linked with climate change. Edge/IoT devices, coupled with AI/ML, can play an important role in preventing and fighting wildfires. Information gathered from environmental sensors deployed in the forest not only offers better monitoring but also helps to predict, detect, and manage wildfires. By using a traditional cloud-centric model, near-real-time analytics on the behavior and spread of wildfires cannot be achieved effectively due to the large amount of information to be transmitted. Improving the data processing capabilities of edge applications that are closer to the response teams deployed on the ground can provide a powerful tool for real-time assessment of wildfires.
We have developed an innovative sensor which is capable of continuous monitoring of IR temperature with a 768-pixel image for flame detection. It is able to capture a 3MP RGB image under flame trigger and report data on CO2, PM1, PM2.5 and PM10 concentrations and the air quality index. Data are transmitted by NB-IoT LTE and CAT-M2 bands to a cloud server for alarm verification and fire time evolution. Edge AI algorithms are used to detect the onset of the flame. Field tests show the ability to detect a flame at a 90m distance with approximately 50 x 50 x 50 cm flame dimensions. The system has been designed to run with ultra-low power processors and electronics with a battery power supply lasting 6 months. A low-cost design for industrial production was also considered for the potential of large-scale deployments.
How to cite: Renzi, F., Coppola, V., Cerreta, R., and Valentini, R.: An innovative IoT system for wildfire detection, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21013, https://doi.org/10.5194/egusphere-egu26-21013, 2026.
Climate change is increasing the frequency and intensity of extreme events like heat waves, droughts and storms, placing forests under growing physiological and mechanical stress.
Common indicators of tree stress, such as sap flow, stomatal conductance, water potential or photosynthetic activity, provide valuable insights but are costly, maintenance-intensive and difficult to scale for continuous, long-term observation. We propose a novel alternative approach: tracking tree sway through its seismic ground motion signature, referred to as the tree's seismic fingerprint. These wind-induced sway signals are intrinsically linked to the mechanical properties of leaves, branches and trunks, which change under environmental stress. Seismometers offer key advantages: they are non-invasive, low-maintenance and easily scalable for tree monitoring across forest plots.
Using observations from ground-based seismometers and trunk-mounted accelerometers at the ECOSENSE site in the Black Forest, we isolated and analysed tree sway signals based on spectral decomposition and vibrational mode tracking. We identified consistent tree-dependent sway frequencies around 0.2 Hz and demonstrated that ground-based sensors can capture sway dynamics without direct attachment. Using machine learning, we further showed that wind speed can be reliably predicted from seismic features, revealing that wind-induced mechanical input is encoded in ground motion.
These findings show that seismometers can passively monitor both environmental forcing and tree biomechanical response. As such, seismic sensing offers a powerful, scalable tool for forest monitoring, with the potential to capture both structural stability and stress-related changes under climate extremes.
How to cite: Umlauft, J., Mora, K., Limberger, F., Gerberding, K., Wirth, C., Werner, C., and Kattenborn, T.: The seismic fingerprint of wind-induced tree sway , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2935, https://doi.org/10.5194/egusphere-egu26-2935, 2026.
The Integrated Carbon Observation System (ICOS) is a European Research Infrastructure (RI) with the goal is to provide high quality greenhouse gasses data collected according to standard protocols and distribute them in real-time with an open data access according to FAIR principles through its dedicated data portal. The terrestrial component of ICOS consists of a distributed network of monitoring stations across Europe covering the most representative ecosystem types (forests, grasslands, croplands, mires,lakes and urban). These stations provide a broad, high-standard ecological assessment of the target ecosystem: measurements are based on the eddy covariance technique from highly equipped flux towers, while a large set of meteorological variables and vegetation related parameters like Leaf Area Index, biomass that are collected according to strict protocols.. The network is currently fully operational with >100 stations delivering data across 16 countries.
This presentation focuses on the potential of the ICOS for serving as a reference in-situ network for satellite product Cal/Val activities, highlighting how the ICOS ecosystem network can be optimised for the CalVal community by adding new sensors and technologies for new variables like Land Surface Temperature (LST), optimising the network of below canopy PAR sensors for estimating fAPAR in forest ecosystems or using Terrestrial Laser Scanning (TLS) for characterizing canopy structure and forest biomass. We also highlight how the cal/val community in particular can contribute through the design, discussion and implementation of specific protocols to match the possible requirements.
How to cite: Gielen, B., Sabbatini, S., Op de Beeck, M., Nicolini, G., and Papale, D.: Optimising the ICOS ecosystem network for satellite cal/val activities by adding new sensors and technologies. , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20737, https://doi.org/10.5194/egusphere-egu26-20737, 2026.
Accurate prediction of carbon exchange in the Central Indian Himalaya, a global biodiversity hotspot with extreme vertical gradients and monsoon variability, remains critical for regional carbon assessments. Despite substantial sequestration potential in Himalayan pine forests, mechanistic drivers of net ecosystem exchange (NEE) are poorly constrained, while alpine grassland carbon dynamics remain enigmatic due to observational scarcity at high elevations. Here, we integrate bidirectional long short-term memory (BiLSTM) networks with SHAP explainability to predict hourly NEE and quantitatively rank environmental controls across contrasting ecosystems. Continuous eddy covariance data from a needleleaf forest (Kosi-Katarmal, 1217 m; April-October 2020-2022) and alpine grassland (Darma Valley, 3240 m; July-October 2022-2023) were analyzed using air temperature, relative humidity, net radiation, soil conditions, vapor pressure deficit, NEE derivatives, and multi-scale lag features (1-24 h). The BiLSTM model achieved exceptional performance (forest: R² = 0.94-0.95, RMSE = 2.18-2.86 μmol m⁻² s⁻¹; grassland: R² = 0.95-0.96, RMSE = 1.09-1.73 μmol m⁻² s⁻¹). Critically, SHAP attribution unveiled fundamentally divergent control mechanisms: forest NEE was governed by rapid temporal dynamics (NEE derivative, SHAP: 0.70) and radiation-temperature coupling (SHAP: 0.02 each), signifying energy-driven photosynthetic control. Conversely, grassland NEE exhibited strong short-term memory (1-h lag, SHAP: 0.35) and atmospheric constraint dominance (temperature, SHAP: 0.06, humidity: 0.03), reflecting stomatal regulation and evaporative demand at high elevation. These findings demonstrate that forest carbon exchange operates as an energy-limited, dynamically responsive system, whereas grasslands function as atmospheric-demand limited systems with pronounced temporal persistence. Our results provide a mechanistic framework for ecosystem-specific carbon flux modeling and demonstrate the efficacy of explainable AI for process understanding in data-sparse mountain regions.
How to cite: Lohani, P. and Mukherjee, S.: Prioritizing Ecosystem-Specific Carbon Exchange Drivers in Central Himalayan Forest and Grassland Using Bidirectional LSTM and SHAP Analysis, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17852, https://doi.org/10.5194/egusphere-egu26-17852, 2026.
Quantifying uncertainties in data-driven upscaling of biogenic carbon fluxes is essential for improving our understanding of global carbon cycle processes and for providing robust priors to atmospheric inversion systems. However, most existing global carbon flux products derived from data-driven approaches either provide no uncertainty assessments, or are limited to incomplete sources.
In this study, we developed an ensemble of net ecosystem exchange (NEE) estimates within the FLUXCOM-X framework to systematically quantify uncertainty contributions from multiple sources across the entire carbon flux upscaling workflow. These sources include choices regarding eddy covariance (EC) measurement post-processing, meteorological forcing, predictor set, training data splitting, and machine-learning model. Specifically, we post-processed raw EC data using a Monte Carlo approach that randomly selects the friction velocity (u*) threshold for each site to quantify uncertainty related to the EC measurement. Meteorological uncertainty was represented using a 10-member, 3-hourly ERA5 ensemble. Predictor selection uncertainty was assessed by applying a hybrid genetic algorithm to select multiple “equally good” predictor combinations used to train predictor ensembles. In addition, uncertainties related to site representativeness and model structure were captured through alternative training data splits and by training machine-learning models (i.e., xGBoost, RF, and MLP) with different random seeds. As a result, a large ensemble of spatiotemporally explicit NEE estimates at hourly and 0.05 deg resolution was generated.
We further analyzed the relative contributions of these uncertainty sources to the total spatial and temporal uncertainty of NEE. Results for Europe indicate that predictor selection uncertainty dominates the upscaling uncertainty, followed by training data splitting uncertainty and EC post-processing uncertainty. In contrast, the ensemble spread associated with meteorological forcing and XGBoost models is relatively small, whereas MLP models exhibit substantially larger spread. The total uncertainty of the ensemble is not uniformly distributed across the study region; instead, it exhibits spatial hotspots particularly in Ireland, west of the United Kingdom, and the northern coast of Africa. The same ensemble-based methodology will next be applied globally to quantify and attribute regional NEE uncertainties worldwide.
How to cite: Yang, Q., Walther, S., Nelson, J., Duveiller, G., Hamdi, Z., and Jung, M.: Partitioning NEE Uncertainty Sources Using a FLUXCOM-X Ensemble, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14806, https://doi.org/10.5194/egusphere-egu26-14806, 2026.
Extreme climate events are increasingly altering biosphere–geosphere exchanges, particularly through their impacts on vegetation productivity and carbon uptake. In this study, we evaluate how multi-dimensional climate extremes shape the spatiotemporal dynamics of Gross Primary Productivity (GPP) across India, a region characterized by strong hydroclimatic variability and rapidly changing land–atmosphere feedbacks. We used FLUXCOM-X GPP estimates (0.25°, 2001–2021) and ERA5-based meteorological variables to derive key climate extremes indices including temperature extremes (TX10P, TX90P), vapor pressure deficit extremes (VPDX90P), soil moisture extremes (SMX10P), and extreme precipitation indicators (RX5day, R20mm). Long-term changes in vegetation productivity were quantified using the Mann-Kendall test and Sen’s slope estimator, while wavelet power spectrum (WPS) and wavelet coherence (WTC) analyses were employed to explore dominant periodicities of GPP variability and its coupling with extreme climatic drivers at seasonal, annual, inter-annual, and long-term timescales. Results reveal widespread increasing trends in GPP across India over the past two decades, driven primarily by intensification of monsoon-season productivity. However, seasonal analysis identifies emerging constraints during the dry season, reflecting the increasing dominance of evaporative stress and atmospheric water demand on vegetation functioning. The impacts of extremes are strongly heterogeneous: heat and VPD extremes exhibit a pronounced negative influence on GPP in central and semi-arid zones with limited water availability; cold extremes reduce productivity in northern and northeastern ecosystems with winter-dominated phenology; soil moisture deficits consistently suppress carbon assimilation across all vegetated systems, while wet anomalies provide strong productivity enhancement; and intense precipitation events generally increase GPP by alleviating moisture stress except in temperate northern regions and north-eastern parts of India, where intense flooding and lower temperatures suppress net growth. Dominant annual cycles in croplands highlight strong synchronization with monsoon-driven growing seasons, whereas forests dominated regions demonstrate inter-annual to long-term modes, reflecting deeper rooting strategies, structural inertia, and ecological memory. WTC results indicate that GPP coherence with high temperature, VPD extremes and low soil moisture extremes is frequently strong and negative, suggesting that future warming and humidity stress may offset current productivity gains. Meanwhile, positive coupling with precipitation extremes implies a growing reliance of ecosystem carbon uptake on episodic wet events. Overall, the findings demonstrate that intensifying climate extremes are actively reshaping the carbon cycle in India by altering both the magnitude and stability of vegetation productivity. As land–atmosphere interactions become increasingly governed by high-impact events rather than gradual change, monitoring multi-scale GPP response and ecosystem sensitivity will be crucial for predicting future terrestrial carbon feedbacks and supporting climate-resilient land management strategies in vulnerable tropical and subtropical regions.
How to cite: Mutyala, P. and Ghosh, P. S.: Climate Extremes Reshape Carbon Uptake in India: A Multi-Scale Assessment of Vegetation–Climate Interactions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-635, https://doi.org/10.5194/egusphere-egu26-635, 2026.
The GOSAT satellite is in orbit since 2009 and now allows a robust evaluation of interannual variability (IAV) in land–atmosphere exchange with global coverage over a period of 15 years. We use monthly net land CO2 fluxes for 2009–2024 inferred from assimilation of GOSAT XCO2 together with in-situ CO2 data in the global inversion system TM5-4DVAR to provide a global overview of IAV of different regions and an in-depth understanding of the long-term carbon cycle over Australia.
At the global scale, we compare the net ecosystem exchange (NEE) based on TM5-4DVAR to the ensemble mean of dynamic global vegetation model (DGVM) estimates from TRENDY v13. We find that DGVMs typically exhibit weaker IAV than XCO2 inversion-based fluxes suggesting that the modelled sensitivity of NEE to hydroclimatic variability remains underestimated in DGVMs.
Following the data fusion approach of our previous studies on semiarid ecosystems (Metz et al., 2023, Metz et al., 2025, Vardag et al., 2025), we then analyse the carbon cycle over Australia, where precipitation dynamics strongly control biogenic fluxes. For Australia, the inversion indicates a pronounced sink during La Niña conditions, but also reveals an exceptionally strong sink anomaly in 2022 to 2024. We investigate the origin of these anomalies using sun-induced fluorescence (SIF) and the gross fluxes of selected DGVMs. We find that GPP has increased strongly in 2022 to 2024 and discuss the role of climate and environmental disturbances for this increase.
Overall, the extended satellite record provides a novel opportunity for improving ecosystem parameterizations and finally reducing uncertainty in the global carbon budget.
References:
Metz, E.-M., Vardag, S.N., Basu, S., Jung, M., ... , Butz, A. Soil respiration–driven CO2 pulses dominate Australia’s flux variability. Science, 379, 1332-1335, https://doi.org/10.1126/science.add7833, 2023.
Metz, E.-M., Vardag, S. N., Basu, S., Jung, M., Butz, A.: Seasonal and in terannual variability in CO2 fluxes in southern Africa seen by GOSAT. Biogeosciences, 22, 555–584, https://doi.org/10.5194/bg-22-555-2025, 2025.
Vardag, S. N., Metz, E.‐M., Artelt, L., Basu, S., Butz, A. (2025). CO2 release during soil rewetting shapes the seasonal carbon dynamics in South American Temperate region. Geophysical Research Letters, 52, https://doi.org/10.1029/2024GL111725, 2025.
How to cite: Vardag, S. N., Basu, S., Mandel, E., Grauer, R., Metz, E.-M., and Butz, A.: A 15-year XCO2-based assessment of the terrestrial carbon-cycle , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10652, https://doi.org/10.5194/egusphere-egu26-10652, 2026.
The active root zone storage (aSrz) is a critical yet unobservable quantity in the water cycle. It represents the dynamic component of subsurface water that can be accessed by ecosystems for evapotranspiration (ET), directly linking water, energy and carbon exchanges across the land-atmosphere interface. In this study, we propose a catchment-scale, ecosystem-oriented hybrid model to understand the spatiotemporal dynamics of aSrz. We introduce aSrz as a latent soil moisture state in the model; without explicitly prescribing soil layers or providing rooting depth information, the model diagnoses aSrz (and its capacity) by relying on first-order ecohydrological principles and multi-source observational constraints (ET, runoff, snow and terrestrial water storage). We train the model across hundreds of U.S. catchments over the period 1985–2020 and then upscale to a 0.25° grid, finding that the inferred root zone storage peaks in transitional regions. We explore the interplay of vegetation, atmospheric demand and water supply, seasonality and topography in modulating the root zone storage dynamics. Furthermore, aSrz capacity reveals the long-term ecosystem adaptation to hydroclimate and substrate conditions. By investigating the differences between aSrz dynamics and its capacity across catchments, we uncover divergent ecosystem strategies for managing water resources, especially along the aridity gradient. Overall, our parsimonious hybrid model structure provides a physically consistent and observationally constrained roadmap for diagnosing ecosystem processes that cannot be directly observed.
How to cite: Blougouras, G., Jiang, S., Brenning, A., Migliavacca, M., Slater, L., Zhou, J., Kumar, R., Tian, L., Wang, C., and Reichstein, M.: Hybrid modeling of the active root zone storage and its capacity, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-940, https://doi.org/10.5194/egusphere-egu26-940, 2026.
Enhanced photosynthetic CO₂ uptake (gross primary productivity; GPP) by terrestrial plants in response to rising atmospheric CO₂ concentrations constitutes the largest and most uncertain ecosystem feedback on climate. Direct measurement of GPP at scales above the leaf is not possible due to co-occurring respiratory fluxes. Carbonyl sulfide (COS) has emerged as a promising tracer of GPP across scales because of its predominantly one-way flux into leaves and the use of leaf relative uptake (LRU) to convert COS uptake into GPP. However, differences in the relative uptake of COS to CO₂ between C3 and C4 vegetation across scales must be accounted for in the application of COS as a tracer and could potentially be applied to detect climate-driven shifts in C3/C4 vegetation distributions. These aspects have largely been ignored, with most COS measurements focusing on C3 vegetation, and only limited C4 COS measurements available.
Here, we develop an atmospheric-based approach to identify and quantify C4 vegetation contributions to large-scale photosynthetic uptake using existing COS and CO₂ concentration measurements from multiple NOAA Global Monitoring Laboratory (GML) network sites. We analyze the seasonal cycle amplitudes of COS as a function of CO₂, defined as atmospheric relative uptake (ARU), and identify sites that deviate from the regression line describing the majority of sites. We derive an analytical framework linking leaf-, ecosystem-, and atmospheric-scale relative uptake, explaining how shifts in physiological traits characteristic of C3 and C4 vegetation influence atmospheric COS and CO₂ signals. Using this framework together with gridded fluxes from the land surface model: Simple Biosphere Model (version 4; SiB4), we show that sites influenced by C4 vegetation exhibit systematic deviations in ARU relative to predominantly C3-influenced sites. These results demonstrate that ARU provides a viable means of detecting and quantifying C4 vegetation contributions to GPP at regional scales (102 – 103 km2), and can be used to detect climate-driven shifts in C3/C4 distributions. Our study advances the use of COS as a tracer of GPP and of C3 and C4 photosynthesis across scales.
How to cite: Bohak, Y. L., Miller, J. B., Montzka, S. A., Rastogi, B., Kaushik, A., and Yakir, D.: Using Atmospheric COS–CO₂ Seasonal Amplitude Ratios to Quantify C4 Contributions to GPP, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17114, https://doi.org/10.5194/egusphere-egu26-17114, 2026.
Accurate quantification of global terrestrial gross primary production (GPP) is critical for understanding the carbon cycle, yet significant discrepancies persist in current estimates regarding their magnitude and spatiotemporal patterns. Solar-induced chlorophyll fluorescence (SIF) has emerged as a promising proxy for GPP; recent mechanistic light reaction (MLR) theories have successfully elucidated the mechanistic SIF-GPP link, while their applicability at the global scale remains unclear. Here, we propose an improved mechanistic light reaction model, qMLR, designed to apply the mechanistic SIF-GPP relationship for global GPP estimation. Building upon the original leaf-scale MLR theory, the model integrates GOSIF with flux tower-based parameter calibration; by implementing a climate-zone-specific calibration strategy based on the Köppen-Geiger system and employing Genetic Algorithms and Bayesian Optimization, we precisely characterized the nonlinear responses of maximum quantum yield of photochemistry (ΦPSIImax) and the fraction of open PSII reaction centers (qL) to environmental gradients, factors previously unaccounted for in global SIF-based GPP estimations. This approach generated a global 0.1° monthly GPP product for the period of 2004-2024. Validation against 425 eddy covariance sites demonstrates that qMLR matches or outperforms existing benchmark products in overall accuracy (R2=0.70), with a regression slope (0.93) closer to unity. The model's mechanistic framework corrects the systematic underestimation prevalent in traditional (e.g., FLUXCOM GPP and MODIS GPP) models over tropical regions: across 7 tropical forest validation sites, qMLR achieved a mean bias of -3.29%, markedly outperforming other mainstream products (mean bias of -28.21%). Our results reveal a global multi-year average GPP of approximately 152.03 ± 4.42 PgC yr-1, higher than the conventional estimate of ~120 PgC yr-1, and show an increasing trend of 0.642 PgC yr-2. This study successfully brings MLR model to global scale and provides a long-term global GPP dataset based on MLR model for the first time, highlights the central role of tropical forests in the global carbon cycle, and offers a new physical benchmark for accurately assessing global vegetation productivity.
How to cite: Zhang, F., Ma, H., Zhang, Y., Lu, Y., Wang, S., and Han, J.: Improved Estimation of Terrestrial Gross Primary Production Using a Mechanistic Light Reaction Model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4762, https://doi.org/10.5194/egusphere-egu26-4762, 2026.
Accurate monitoring of terrestrial CO₂ uptake is essential for Natural Climate Solutions and reducing carbon accounting uncertainty. Project-scale certification protocols can provide robust estimates but often depend on costly site-level measurements and are difficult to scale. Global carbon flux models provide continuous coverage, but their coarse resolution cannot represent heterogeneous land management. Bridging these approaches requires high-resolution, scalable carbon monitoring with transparent uncertainty estimates.
Within the BenchFlux project, we present initial results from a High-Resolution Carbon Flux Monitoring framework based on the Global Estimation Model (CARBON-GEM). CARBON-GEM integrates (i) surface reflectances from the HISTARFM data-fusion approach (Moreno-Martinez et al., 2020), (ii) meteorological drivers, and (iii) eddy-covariance (EC) observations to estimate both gross primary production (GPP) and net ecosystem exchange (NEE) daily at 30 m resolution. The approach utilizes machine-learning methods, such as neural networks, to capture nonlinear responses, and is implemented in Google Earth Engine for scalable mapping. The workflow also delivers pixel-level uncertainty quantification, moving beyond categorical quality flags to support auditability and interpretation.
In addition to standard out-of-sample cross-validation to assess robustness and generalization, we validate CARBON-GEM against independent, scale-aware FluxMapper ground truth (Metzger, S., 2018) provided by BenchFlux’s SpatialEddy component. FluxMapper couples next-generation EC processing with flux spatialization to enable explicit space-time matching and local-to-regional nesting. In this context, CARBON-GEM extends the FluxMapper-scale structure beyond individual stations, allowing continuity across diverse landscapes. Complementing this, FluxMapper provides a novel, independent benchmark for high-resolution carbon-flux estimates and serves as a robust reference point, alleviating standard EC spatial-resolution constraints and facilitating the decomposition of aggregate point measurements into fine-grained spatial patterns. CARBON-GEM and FluxMapper together establish a foundation for scalable, uncertainty-aware 30-meter monitoring of GPP and NEE. This approach captures essential spatial heterogeneity necessary for large-scale real-world auditing in NCS planning, reporting, and verification.
- Moreno-Martínez, Á., Izquierdo-Verdiguier, E., Maneta, M. P., Camps-Valls, G., Robinson, N., Muñoz-Marí, J., ... & Running, S. W. (2020). Multispectral high-resolution sensor fusion for smoothing and gap-filling in the cloud. Remote Sensing of Environment, 247, 111901.
- Metzger, S. (2018). Surface-atmosphere exchange in a box: Making the control volume a suitable representation for in-situ observations. Agricultural and Forest Meteorology, 255, 68-80.
How to cite: Moreno-Martinez, A., Izquierdo-Verdiguier, E., Muñoz-Mari, J., Hirn, J., Endsley, A., Desai, A., Metzger, S., Bower, S. J., Robinson, N., Greenberg, S., Clinton, N., and Camps-Valls, G.: From Global to Local: Precision Carbon Flux Mapping for Natural Climate Solutions, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12874, https://doi.org/10.5194/egusphere-egu26-12874, 2026.
The net ecosystem exchange of CO2 (NEE) between the land and the atmosphere is a critical term in the global carbon budget. Because of the complexity of modeling NEE across scales, global estimates of NEE are subject to large uncertainties. The two major data-driven approaches to modeling NEE are commonly described as top-down and bottom-up. Top-down models create a estimate of NEE which is optimally consistent with observations of atmospheric CO2 from tower, aircraft, and increasingly satellite sensors. Bottom-up NEE models learn a statistical relationship between a set of ecosystem-level biophysical drivers and observations of NEE, often from the global eddy-covariance network. These models are then upscaled to the globe using remotely sensed data. These systems are critical to earth system science. However, both are subject to limitations and disagreement based on the particular view which they represent.
In previous work we presented two frameworks for integrating top-down and bottom-up approaches. Both studies build a data-driven bottom-up NEE model trained from eddy-covariance data, which is also constrained by atmospheric information. The atmospheric constraint in the first study, derived statistically from an ensemble of atmospheric inversions, created a model which strongly adjusted regional and global model results towards top-down and independent estimates of NEE, albeit with limited improvement in the model’s spatial and temporal representation of NEE. The atmospheric constraint in the second study, derived from direct observations of atmospheric CO2 using a Lagrangian atmospheric transport model, improved the representation of NEE in biomes which are under represented in the eddy-covariance record. This resulted in an improved representation of the dynamics of NEE, producing spatial and temporal variability which better represents independent estimates and our current ecological understanding. However, the second atmospheric constraint produced a model with high internal uncertainty, and which underperformed at the regional and latitudinal scale, producing less plausible annual timeseries and mean seasonal cycles when compared with other bottom-up data-driven models.
In the current work, we present a model which combines these two constraint techniques: The model uses 1) a core constraint from eddy-covariance, 2) A statistical constraint from atmospheric inversions to limit the possible solutions, reducing uncertainty, and improving regional results, and 3) an atmospheric constraint from direct observations of atmospheric CO2 to improve the representation of the regional and global dynamics of NEE. Using the three constraints in parallel, the new model produces an estimate of global NEE which preserves the strengths of the two previous studies. When compared with state-of-the-art bottom-up models, it produces improved regional results, consistent spatial and temporal dynamics, and lower internal uncertainty. When transported through the atmosphere, the new model produces realistic estimates of atmospheric CO2. In this way, we demonstrate the progress towards a mature hybrid framework, which can inherit the strengths of both bottom-up and top-down approaches.
How to cite: Upton, S., Reichstein, M., Peters, W., Botia, S., van der Woude, A., Nelson, J. A., Walther, S., Jung, M., Gans, F., Haszpra, L., and Bastos, A.: Towards a mature framework for integrating bottom-up and top-down constraints in a data-driven ecosystem-level CO2 model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18972, https://doi.org/10.5194/egusphere-egu26-18972, 2026.
Posters on site: Wed, 6 May, 08:30–10:15 | Hall X1
The water cycle in forests of many regions is being impacted by climatic changes, often including a decrease in precipitation and an increase in temperature, leading to an increase in green water (evapotranspiration) and a decrease in blue water (runoff + drainage). Additionally, forest expansion and development are prevailing processes in many rural areas due to the abandonment of traditional land uses. Stand leaf area growth further amplifies green water and reduces blue water availability. However, the interaction between climate change and stand structure changes is not well understood at large scales. We modeled Spanish forest water cycle using national inventories (1990–2020), analyzing climate and forest structure trends at plot level. Using three inventory surveys, we assessed green and blue water changes across forest types and compared managed vs. unmanaged forests. Results show green water increased and blue water decreased over time. Leaf area index (LAI) growth (trees and shrubs) had a stronger effect on green water than climate change. These factors, along with recent precipitation declines (2010–2020), also significantly reduced blue water. Basal area reduction improved blue water yield, but only in the short- mid-term, as stand LAI tended to recover over time. This study demonstrates that changes in stand structure can be as important, if not more so, than climatic changes in influencing the water cycle at the regional level. Moreover, our results support the idea that effective basal area reduction can enhance blue water production, but only if basal reduction practices are consistently maintained.
How to cite: Sánchez, J., De Cáceres, M., Vayreda, J., and Retana, J.: Recent water cycle changes in Spanish forests are driven by stand structure more than climatic changes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-4304, https://doi.org/10.5194/egusphere-egu26-4304, 2026.
The BenchFLUX project represents an important advance in evaluating nature-based climate solutions (NbCS) to address the growing climate crisis. The benchmarking of CO₂ fluxes using flux tower measurements and Earth Observation (EO) data is the project's aim, employing multiple approaches to introduce, compare, and integrate temporal and spatial scales. The methods used account for the nonlinear behavior of carbon flux dynamics across scales. Therefore, measurement harmonizations are fundamental for aligning ground and atmospheric measurements. And thus, BenchFLUX provides reliable models and products that accurately track carbon emissions from small local areas to the global scale.
To achieve this goal, the project combines eddy covariance flux tower ground data with multi-source EO data to create harmonized datasets for various advanced machine learning models at different scales. The processes use cloud computing technologies, such as Google Earth Engine and cloud-optimized workflows, to produce spatial CO₂ flux data at multiple spatial resolutions. The proposed methods, including Bayesian and knowledge-guided approaches to achieve accurate and consistent results, and the final products are nested across different temporal and spatial scales among the six international research teams, serving as an integrated element for cross-scale continuity.
The spatial scalability of these methods is analyzed in the project prototype results. Preliminary monthly average CO₂ exchange (GPP) results are provided from the highly standardized NEON sites database for the higher spatial resolution models, revealing discrepancies at multiple scales during both the growing and non-growing seasons. The initial results will also compare coarser spatial resolution models with the eddy covariance ground truth data. All these ongoing comparisons aim to identify the most reliable methods for scaling carbon flux estimates. This will help determine the best combination of techniques to ensure high local precision and global consistency, ultimately supporting continuous cross-scale resource management.
How to cite: Izquierdo-Verdiguier, E., Moreno-Martinez, A., Stoy, P., Sonnentag, O., Pal, C., Kang, Y., Keenan, T., Desai, A. R., Metzger, S., Fortier, M., Bassiouni, M., Ranjbar, S., Bower, S., Hoffman, S., Losos, D., and Xiao, J.: Reconciling Cross-Scale Discrepancies in CO₂ Fluxes. Preliminary Findings from the BenchFlux Project, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10567, https://doi.org/10.5194/egusphere-egu26-10567, 2026.
Compound climate extremes are projected to increase in both intensity and frequency under future climate scenarios. While atmospheric heatwaves are well-documented, recent evidence suggests that soil temperature extremes can be more persistent and prominent than air temperature, leading to devastating compound soil heat and drought (CSHD) events. Despite their potential severity, our understanding of how these soil compound extremes impact terrestrial vegetation productivity remains limited.
This study utilizes high-frequency datasets from a global network of eddy covariance towers—including AmeriFlux, FLUXNET-2015, ICOS, and JapanFlux—to quantify the impact of CSHD on Net Ecosystem Productivity (NEP). We employ a data-driven machine learning framework to isolate and estimate the productivity losses specifically attributable to the compound soil heat and drought stress. We move beyond and use an explainable machine learning approach (XAI) to identify the primary drivers and reveal the non-linear sensitivities of various ecosystems to these extremes. Our findings provide critical insights into the resilience of the terrestrial biosphere and improve our ability to predict ecosystem responses to increasingly complex climate stressors.
How to cite: Shekhar, A.: Impact of compound soil heat and drought on terrestrial vegetation productivity, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13243, https://doi.org/10.5194/egusphere-egu26-13243, 2026.
Extreme hydrometeorological events have become more frequent and intense, with direct implications for carbon, water, and energy exchanges between the biosphere and the atmosphere. Decades ago, extensive areas of natural wetlands in southern Brazil were converted into flooded rice paddies. The extreme floods that affected the state of Rio Grande do Sul in 2024 highlighted the critical regulatory role of these areas. Even when used for agriculture, wetlands acted as natural hydrological buffers, attenuating flood peaks. However, prolonged inundation can substantially alter greenhouse gas exchanges, particularly methane (CH₄), under anaerobic soil conditions. This study quantified the impacts of the extreme floods of 2024 on ecosystem–atmosphere exchanges of CO₂, CH₄, and H₂O, based on continuous eddy covariance measurements conducted in an irrigated rice lowland in southern Brazil. The site is managed under a typical intensive regional system and followed the crop rotation sequence: flood-irrigated rice (December 2023 to April 2024); winter fallow due to prolonged flooding (May to October 2024); rainfed soybean (December 2024 to April 2025); and forage crops with cattle grazing (May to October 2025). The study compared the May–October period of 2024 (flood year) with the same period in 2025 (non-flood year). During the prolonged inundation period in 2024, the system exhibited higher CO₂ and CH₄ emissions compared to the corresponding non-flooded period in 2025, while Evapotranspiration was similar. The absence of flooding and the cultivation of forage crops in 2025 resulted in reductions of up to 20% in CO₂ emissions and 60% in CH₄ emissions relative to the flooded fallow period of 2024. These results demonstrate that extreme hydrological disturbances can induce short but intense pulses of greenhouse gas emissions, with persistent effects on annual carbon balances. At the same time, adaptive management practices, such as crop rotation and the reduction of fallow periods, show potential to mitigate these effects and enhance agroecosystem resilience. The findings contribute to the derivation of local emission factors, the development of climate-adaptive agricultural strategies, and integrated assessments of extreme events at local and regional scales.
How to cite: Roberti, D. R., Mergen, A., Maboni, C., Zimmer, T., Oliveira, M. E., dos Santos, D. M., de Souza, E. C., Lopes, M., Stefanello, M. B., Basso, J. V., Sathres, V. L., da Silva, J. V., da Rosa, H. V. B., and Jacques, R. R. J.: Impacts of the 2024 Extreme Flooding Disaster on Greenhouse Gases Exchanges in Rice Floodplains of Southern Brazil, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13745, https://doi.org/10.5194/egusphere-egu26-13745, 2026.
The frequency and intensity of extreme weather events (EWEs) have increased in recent decades and are projected to continue rising globally. The impact of EWE on boreal ecosystems can be disproportionate. In forests, the exchange of CO2 is affected by droughts depending on their timing and severity. In lakes, heatwaves might strengthen stratification and trigger oxygen depletion, with consequences on greenhouse gas (GHG) dynamics. Excessive heat and winds can also increase GHG emissions from peatlands and lakes. The driving mechanisms of GHG exchange in these three ecosystems differ, and their responses to EWE also vary.
Long-term flux data, obtained with the Eddy covariance (EC) technique, enable us to establish a baseline response and analyse how these different ecosystems react to EWE. The EC technique measures the vertical exchange of gases, particles, and energy at an ecosystem scale, with a 30-minute interval, demonstrating the instantaneous response of the ecosystem.
Here, we analysed the effect of EWE on GHG dynamics from adjacent forest (Hyytiälä), lake (Kuivajärvi) and peatland (Siikaneva) ecosystems, where long-term EC flux measurements are available. EWE, such as heatwaves, dry spells, excessive rainfall, prolonged high wind spells, and compound events, have been identified in the last decade using both in situ and ERA5 reanalysis Land hourly datasets. The ecosystems exhibit contrasting responses to the EWEs. For instance, during the 2018 heatwave, the forest exhibited enhanced CO2 uptake, while both the lake and the peatland showed increased emissions relative to the reference period (2013-17).
How to cite: Thomas, A., Gunnlaugsdóttir, E. G., Fregona, M., and Mammarella, I.: Extreme weather event responses of collocated forest, lake and peatland ecosystems, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14463, https://doi.org/10.5194/egusphere-egu26-14463, 2026.
Accurate representation of forest structure and above-ground biomass (AGB) at the individual-tree scale is essential for improving carbon cycle assessments and enabling emerging forest Digital Twin applications. Terrestrial Laser Scanning (TLS) provides high-resolution three-dimensional observations of vegetation structure; however, transforming dense point clouds into biologically meaningful and computationally efficient tree-level models remains challenging. Here, we present a TLS-based end-to-end framework for individual-tree structural reconstruction and component-wise AGB estimation using high-density point cloud data acquired over a small forest copse within the Indian Institute of Technology Kharagpur campus using a Terrestrial Laser Scanning system.
The workflow begins with multi-scan point cloud registration and preprocessing, including noise removal, ground filtering, and vegetation isolation. Multi-view acquisition is employed to mitigate occlusion, and residual data gaps are addressed through model-based structural interpolation of woody elements. Individual trees are delineated using geometric clustering-based instance segmentation (TreeIso), with segmentation quality assessed against field-mapped stem locations. Leaf and woody components are separated using a combination of graph-based structural analysis. Woody architecture, including stems and branches, is reconstructed using quantitative structure modelling (TreeQSM), where adaptive cylinder fitting is applied to derive branching topology and woody volume. Leaf biomass is estimated independently by converting classified leaf points into a voxel-based crown representation from which leaf area is derived. Leaf mass is then calculated using species-specific specific leaf area (SLA) values obtained from field sampling. Species identity and corresponding wood density values are assigned using concurrent field inventory data. Component-wise woody and foliar masses are combined to obtain tree-level AGB estimates. Each stage is implemented using alternative models and parameterisations. Selected models are regionally calibrated to improve performance under the conditions of the study area.
To ensure biological realism, reconstructed tree geometry is validated against field measurements and reference allometry using stem diameter at breast height (DBH), tree height, and crown metrics, and sensitivity analyses are conducted to quantify uncertainty propagation from segmentation, classification, and QSM parameterisation into final biomass estimates. The final framework demonstrates the potential of TLS-derived point clouds to produce validated, structurally explicit tree-level models that support carbon accounting, ecosystem modelling, and calibration of airborne and satellite-based biomass products, thereby bridging in situ measurements and multi-scale Earth observation systems.
Keywords : Terrestrial Laser Scanning (TLS); Individual-tree segmentation; Leaf–wood separation; TreeQSM; Above-ground biomass (AGB).
How to cite: Vinesh, B. and Behera, M. D.: A TLS-Based Framework for Individual-Tree Structural Reconstruction and Improved Biomass Estimation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16346, https://doi.org/10.5194/egusphere-egu26-16346, 2026.
Accurately representing terrestrial carbon fluxes from local ecosystems to continental scales is a central challenge for land-ocean interface, particularly concerning the fate of carbon transformed during transportation. It is required to apply robust validation to accurately project continental carbon fluxes, yet the ultimate fate of carbon exported from land to ocean remains a key uncertainty. While marine sedimentary archives provide an integrated, long-term record of this flux, our ability to interpret those after degradation and diagenetic processes is paramount. Fluorescence EEM-PARAFAC is a promising technique for the source identification in water column, but we argue its application in sedimentary organic matter.
The conventional use of EEM-PARAFAC assigns fluorescent components (C, A, M) to specific terrestrial or marine sources, underpinning popular indices like the Humification Index (HIX). However, we show this source-centric view is incomplete for three key reasons. First, the diagenetic environment can overwrite source signals; anoxic processing of terrestrial matter, for example, can mimic a 'marine' low-HIX signature. Second, key indices like HIX are inherently sensitive to organic matter concentration, requiring careful re-calibration to separate measurement artifacts from true biogeochemical change. Third, thermal alteration during deep burial further degrades and distorts these fluorescent signals.
Instead of viewing these sensitivities as confounding factors, we propose repurposing them as diagnostic tools based on the results of indoor incubation monitoring and in-situ profiles combining paleoenvironmental analysis. In this new framework, fluorescent signatures become proxies for the environment itself with correction and background information: HIX, AC/M, and P/H can trace paleoenvironments, e.g., historical redox conditions, paleo-thermometers.
By embracing this holistic approach, we transform EEM spectroscopy from a simple source-tracker into a dynamic environmental recorder. This study unlocks a richer, multi-layered narrative of carbon's journey from source to sequestration, providing a powerful new set of process-based constraints.
How to cite: Gan, S.: Bridging Molecular Signatures from Terrestrial to Continental: A Critical Re-evaluation of EEM-PARAFAC as a Diagnostic Tool for Carbon Source Fingerprints, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17050, https://doi.org/10.5194/egusphere-egu26-17050, 2026.
Global biochemical cycles in forest ecosystems may shift as a result of climate change conditions. It is essential to comprehend how biogeochemical cycles and environmental factors control nutrient releases and storage in soils and forest ecosystems. Moreover, soil is the largest storage pool of carbon and nutrients. Nutrient availability, particularly in litter and soil, directly influences tree growth and biomass accumulation, which in turn impacts forest structure, productivity, nutrient cycling, and soil CO2 emissions.
In this work, we investigate how the litter nutrient dynamics and turnover can influence the soil CO2 emissions in Mediterranean oak-dominated ecosystems with a soil manipulation experiment in the Xanthi region of northern Greece. We studied soil CO2 efflux under three organic matter input treatments: Control [CON] (undisturbed), No-Litter [NL] (aboveground litter excluded) and No-Litter-No-Roots [NLNR] (both litter and roots excluded). Monitoring plots were established in a broadleaf evergreen ecosystem dominated by Quercus coccifera L. and Phillyrea latifolia L. Equal plots were established in a deciduous oak forest dominated by Quercus frainetto Ten., followed by Quercus cerris L. and Quercus petraea (Matt.). Using a Li-8100 automated soil CO2 efflux system, soil respiration, moisture and temperature measurements were conducted at 54 points in total, once every three months for two years (2023-2024). We also analyzed a range of nutrients in litterfall and forest floor and estimated their turnover rates to determine their effect on soil respiration. Our results showed that in both ecosystems, soil temperature and moisture predominantly controlled soil CO2 effluxes. Litter turnover was identified as a key driver of soil CO2 efflux in broadleaf evergreens linked to the nutrient dynamics of carbon (C), nitrogen (N), manganese (Mn), and calcium (Ca). Similarly, in deciduous oaks, litter turnover significantly influenced soil CO2 efflux, particularly in relation to C, N, C/N, and potassium (K). This work supports our better understanding of the influence of nutrient cycling on soil emissions in Mediterranean forest ecosystems.
How to cite: Zacharoudi, S., Spyroglou, G., Fotelli, M., and Radoglou, K.: Litter nutrient turnover influences soil CO2 emissions in oak-dominated ecosystems , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17965, https://doi.org/10.5194/egusphere-egu26-17965, 2026.
Mediterranean cork oak forests are increasingly threatened by abiotic and biotic stressors, with Phytophthora emerging as a major cause of tree decline. This study presents preliminary results from an ongoing monitoring experiment aimed at investigating the effects of Phytophthora infection on hydraulic functioning and growth dynamics of cork oak (Quercus suber L.) in a stand located in central Sardinia (Italy). Two experimental theses were considered: healthy trees and declining trees affected by Phytophthora.
Ten trees were continuously monitored starting from May 2025 using TreeTalker®Cyber devices (five per thesis) to measure sap flow velocity, stem radial growth, and microclimatic variables, including air temperature, relative humidity, and vapour pressure deficit (VPD). High-frequency physiological data were integrated with atmospheric conditions to assess differences in tree water use and growth performance between the two theses.
Preliminary results showed a substantial reduction in sap flow magnitude and altered diurnal patterns in declining trees compared to healthy individuals. In addition, declining trees exhibited a pronounced reduction in stem radial growth compared to healthy individuals, indicating a combined impairment of hydraulic functioning and growth processes associated with Phytophthora infection. These findings demonstrate the potential of IoT-based proximal sensing for detecting early physiological signals of tree decline and support its application in forest health monitoring.
How to cite: Riggi, S., Scanu, B., Salbitano, F., Lo Cascio, M., Spano, D., Valentini, R., and Sirca, C.: Early detection of cork oak decline in Mediterranean forests using TreeTalker®Cyber physiological monitoring, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18587, https://doi.org/10.5194/egusphere-egu26-18587, 2026.
Climate across the globe is continuing to change drastically, and ecosystems are being affected through seasonal and inter-annual climate changes and extreme weather events, with the global averaged temperatures reaching record-breaking highs every year, with 2024 being the warmest on record since 1850.
It is known that forest ecosystems play a key role in mitigating temperature, buffering it within the forest microclimate compared to the macroclimate, thus dampening the effects of extreme temperature conditions. But the extent of the effect of these drivers and macroclimate conditions on microclimate conditions is not well understood.
Macroclimate is defined as the set of meteorological variables on a large spatial scale (up to hundreds of kilometers), whereas microclimate is defined by climatic conditions on small spatial scales that result from the interaction between the macroclimate, and forest and topography factors.
As the climate continues to change, learning which features of microclimates help buffer against intense macroclimatic conditions will be paramount. The aim of this study is to quantify the relative effect of macroclimate conditions, forest structure measures, and topographical variables on the microclimatic conditions, through machine learning with gradient boosting machines, and further, to explore how remote sensing data can be used to predict the buffering capacity of microclimates with future macroclimatic conditions. A pilot test is conducted specifically in a mixed forest in Piegaro, central Italy.
With the use of innovative IoT (Internet of Things) sensors, the temperature, relative humidity, and spectral data for selected trees is measured from underneath the canopy. These microclimatic measurements are used to find relationships with macroclimate and other data sources, including NDVI measurements, hourly climate datasets, downscaled- and projected-hourly climate data, and a digital terrain model (DTM). Utilizing data at different scales, from meters to several kilometers, allows the elements of the climate to be explored at varying resolutions, and the differences between these can further uncover the drivers of microclimatic conditions and the importance of including microclimates within climate studies.
It is well understood that the most influential variables on the microclimatic conditions are the corresponding macroclimatic conditions, and it is expected that elements such as relative elevation and aspect play an influential role in microclimatic buffering. Quantifying these relationships can help improve modelling forecasts that generally make use of climate measured on a larger scale, as they disregard the intricacies of microclimates and the possible effects that microrefugia have on species preservation. With further knowledge of microclimates comes a better understanding of how we can prepare for individually-experienced changes in the climate, in a way that promotes native landscapes as well as conserving biodiversity and enhancing local species.
How to cite: Mannsfeld, A. E., Samad, N., and Chiriacò, M. V.: Exploring Drivers of Forest Microclimates in Central Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20113, https://doi.org/10.5194/egusphere-egu26-20113, 2026.
Gross Primary Production (GPP), the amount of CO2 that plants absorb due to photosynthesis, is the biggest carbon flux between biosphere and atmosphere. The current study investigates the estimation of daily GPP of natural vegetation from new generation vegetation indexes (VIs) and compares their performance to the production efficiency model of MODerate resolution Imaging Spectroradiometer (MODIS), MOD17. Two VIs are considered: the kernel version of Normalized Difference Vegetation Index (kNDVI), and the near infrared reflectance of vegetation (NIRV).
kNDVI exploits the higher order relations between the reflectance in the NIR and red regions by defining NDVI in Hilbert spaces and using the radial basis function reproducing kernel. It uses a length-scale parameter (σ) that can be defined conveniently for a specific purpose. NIRV is the product of NIR reflectance and NDVI. It represents the proportion of pixel reflectance attributable to the vegetation in the pixel.
VIs are calculated from MODIS daily continuous surface reflectance in red and NIR (FluxnetEO dataset version 2). This dataset offers quality checked and gap-filled daily MODIS surface reflectance observations during the 2000-2022 period centered in 647 eddy covariance (EC) sites located around the world. Different linear models are trained using VIs alone and combined with photosynthetically active radiation (PAR) measured at EC sites. Observations from 34 EC sites during the 2016-2020 period are used to optimize regression parameters and σ for three different biomes: grasslands, deciduous broadleaved forests, and evergreen needleleaved forests.
The daily GPP estimates are added to 8-day periods according to MOD17 frequency. The three GPP series are validated against EC observations and their results are compared. Using VIs alone, kNDVI achieved correlation R ϵ [0.79,0.87], relative mean bias error rMBE (%) ϵ [-9,6], and relative root mean squared error rRMSE (%) ϵ [52,60]; NIRVR ϵ [0.79,0.87], rMBE (%) ϵ [-8,7], rRMSE (%) ϵ [52,60]. In combination with PAR: kNDVI R ϵ [0.81,0.88], rMBE (%) ϵ [-9,14], rRMSE (%) ϵ [52,58]; NIRVR ϵ [0.81,0.88], rMBE (%) ϵ [-9,22], rRMSE (%) ϵ [51,61]. MOD17: R ϵ [0.41,0.70], rMBE (%) ϵ [-34,18], rRMSE (%) ϵ [35,56].
How to cite: Sánchez-Ruiz, S., Campos-Taberner, M., Martínez, B., Jiménez-Guisado, A., García-Haro, F. J., and Gilabert, M. A.: Daily GPP of natural vegetation from new generation vegetation indices. Comparison with MOD17, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20952, https://doi.org/10.5194/egusphere-egu26-20952, 2026.
High-frequency dendrometer measurements provide valuable insights into short-term and seasonal tree growth dynamics, enabling detailed analyses of forest responses to climatic variability. At the same time, their practical use is strongly limited by data quality issues. Sensor freezing during cold periods, signal drift, data gaps, and site-specific artefacts introduce substantial noise and uncertainty into dendrometer time series. These effects persist even after expert-based corrections and challenge standard assumptions about the availability of reliable ground truth observations.
In this study, we investigate how data quality and model performance can be evaluated when both predictions and reference measurements are affected by uncertainty. We analyse multi-year, hourly dendrometer records of individual tree radial growth collected at forest monitoring sites in Austria, combined with in-situ environmental variables such as air temperature, precipitation, and soil moisture. As a modelling baseline, we employ statistically grounded time-series approaches, including exponential smoothing and seasonal autoregressive integrated moving average models with exogenous variables (SARIMAX). Lagged environmental predictors are incorporated to capture delayed physiological responses of trees to climatic drivers and to reflect the strong temporal dependencies present in the data.
Rather than focusing exclusively on predictive accuracy, we place emphasis on diagnosing data reliability and understanding how observational noise propagates through time-series models. We show that classical evaluation metrics are often insufficient when the target variable itself is noisy or partially unreliable. To address this, we adapt anomaly detection concepts to the specific characteristics of dendrometer data, developing season-aware diagnostics that help identify implausible growth patterns, abrupt regime changes, and periods of degraded sensor performance while preserving biologically meaningful variability.
In addition, we explore how model-based explanations can support data quality assessment in a diagnostic sense. Feature attribution analyses computed over multi-lag input structures are used to examine when model behaviour is driven by consistent environmental signals and when it becomes unstable or difficult to interpret. Rather than treating explainability as an end in itself, we use these attribution patterns as indicators of potential data issues, such as sensor artefacts or inconsistent environmental responses, that warrant closer expert inspection.
The combined use of anomaly-aware diagnostics and explanation-informed analysis provides complementary perspectives on uncertainty and noise in high-frequency ecological time series. Our results highlight the importance of data-centric evaluation strategies for tree growth modelling and demonstrate that interpretable statistical baselines remain essential tools when working with noisy environmental sensor data. The proposed framework supports more robust and transparent downstream applications, including growth forecasting, stress detection, and climate impact assessment under increasing climatic variability.
How to cite: Disarlo, V., Wachsenegger, A., Lampert, J., and Zolles, A.: Explainable Machine Learning for diagnosing Data Quality Issues in Dendrometer-Based Tree Growth Time Series, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21643, https://doi.org/10.5194/egusphere-egu26-21643, 2026.
Forest monitoring under climate change and extreme events remains constrained by the scarcity of continuous in-situ observations in remote areas introducing spatial bias in reference networks, weakening disturbance attribution, and limiting calibration and validation of Earth Observation (EO) products. The EU project RemoTrees addresses this limitation by developing an autonomous forest monitoring system that combines low-power IoT sensing with satellite communication tested across sites representing contrasting forest biomes.
We present the deployment framework to (i) select monitoring domains and (ii) define harmonized sensor installation on stems, in soils, and for reference radiation measurements under consistent criteria across sites. At Level-1 sites, prototypes are benchmarked against established reference instrumentation and protocols to quantify accuracy, drift, reliability, and sensitivity to environmental conditions. Resulting performance and calibration diagnostics guide the transfer to Level-2 sites, which are more remote and demanding, evaluating operational robustness and calibration stability under extreme conditions, including sensor nonlinearities, power and performance limitations, and communication reliability.
Spatial sampling is formalized through a hierarchical hexagonal tessellation based on H3 indexing. Nested hexagons define spatial units from the installation scale to the domain scale, generating an internal network with homogeneous spatial coverage and explicit neighbourhood relations and, by providing scale-compliant aggregation units aligned with EO pixel footprints and uncertainty cores, enable statistically comparable in situ–EO matching beyond single-point validation. This H3-based spatial framework links point measurements with EO products at pixel resolution and enables coherent aggregation from tree to domain while reducing reliance on single-point observations. Domain selection and installation location choice are driven by a multi-criteria spatial decision analysis that integrates: (1) crown-scale structural phenotyping and geo-environmental covariates derived from airborne laser scanning surface models; and (2) Sentinel-2 space–time data cubes aggregated to individual crown objects, including spectral indices and biophysical/-chemical variables i.e. fAPAR, LAI, leaf chlorophyll, and canopy chlorophyll content selected for their functional linkage to photosynthesis, radiation use efficiency, and water stress as well as relatability to in situ measurements of radiation, sap flow, and radial growth. Decisions are implemented as a two-step process: minimum suitability filtering followed by a weighted, normalized composite ranking applied consistently to both domains and candidate crowns/trees, to capture intra-domain variability and optimize final tree and soil-point selection.
The resulting design turns sensor placement into an explicit, transferable multi-scale sampling scheme supporting continuous time series of key variables, including fAPAR; multispectral measurements of incident, canopy reflected and transmitted solar radiation, tree stem radial growth and motions, sap flow, soil temperature and moisture, besides air temperature and humidity below and above the forest canopy. In parallel, the workflow consolidates a traceable inventory of the monitoring domain and instrumented locations, structured as metadata for database integration and analytical use, thereby supporting cross-site comparability and transfer of the design to Level-2 deployments under FAIR principles, and interoperability between in situ and EO systems.
How to cite: Mestre Runge, C., Reudenbach, C., Andreatta, D., Belelli Marchesini, L., Brede, B., Camacho, F., Löening, D., Lotz, C., Valentini, R., Vescovo, L., and Opgenoorth, L.: An innovative experimental design based on Uber Hexagons for strategic IoT sensor placement across contrasting remote forest ecosystems: a proposal from the RemoTrees project, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14420, https://doi.org/10.5194/egusphere-egu26-14420, 2026.
Stubble burning has become a recurrent event in many parts of India, including the central Indian state of Madhya Pradesh. The farmers burn the crop residue in the post-harvest period of the Kharif and Rabi seasons, to quickly prepare fields for the next cropping cycle. This study used MODIS satellite data-derived burned area product and Sentinel-5P derived NO2 column number density to identify fire burned areas and assess the corresponding GHG emission, respectively. The uncertainty layer was utilized in the MODIS data pre-processing, followed by monthly data aggregation. The ESRI land use land cover (LULC) layer was used to differentiate the burned areas in cropland and forest. The study observed a total of 5787 sq km fire-affected area in 2024, wherein crop residue burned area was 5346 sq km that was more than 12 times the forest fire area (441 sq km). More than 75% of the forest fire burning occurred in March, April, and May; with around 15% in post-monsoon November and December. Similarly, more than 74% of the crop residue burning occurred in March, April, and May; and around 20% in post-monsoon November. Overall, the maximum total fire events occurred in April (42%), followed by March (18%) and May (15%). On the contrary, the maximum NO2 concentration was recorded in May and June, followed by November and December. The study hypothesizes that the temporal lag in NO2 concentration relative to the total burned area may indicate the accumulation of NO2 along with the contributions from neighboring states with higher anthropogenic GHG emissions. Moreover, the forest dwellers in central India often apply surface fire for forest floor cleaning during the peak summer months (May and June) before minor forest product collection. These under canopy surface fire events often remain undetected due to dense forest canopy, although they emit a significant amount of GHG. Thus, the GHG emissions from surface fire events remain unaccounted for. The study proposes developing a citizen science-based surface forest fire monitoring module. The accurate canopy and surface fire events, and the crop residue burned area would help assess GHG emission. Such as robust data can be used to develop predictive models for future surface fire risk estimation and quantification of the associated GHG emissions.
How to cite: Das, P., Debburman, P., Behera, M. D., and Chowdhary, V. M.: Assessing GHG Emissions from Forest Fire and Stubble Burning using Satellite Remote Sensing, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1073, https://doi.org/10.5194/egusphere-egu26-1073, 2026.
Forests exert a strong buffering effect on below canopy air temperature, lowering daytime maxima and increasing winter minima. Moreover, forest composition and structure deeply condition the microclimate within and above the forest environment, affecting all its living organisms and their biological processes. Standard meteorological measurements of air temperature—according to WMO recommendations—are not performed under forest cover, thus not adequately representing the temperature regimes driving processes such as photosynthesis, respiration, transpiration, and which serves as environmental cues in regulating leaf and growth phenology. Current forest biogeochemical models rely on gridded temperature data derived from standard meteorological stations and satellite-derived land surface temperatures as drivers to simulate how carbon, nitrogen and water cycles are affected by climate change, potentially leading to biased estimates. Recent availability of low-cost IoT temperature sensors enables the development of dense networks within forest environments, making multi-year, multi-site, high-spatiotemporal resolution microclimate monitoring increasingly feasible.
In this ongoing study focusing on six sites in northeastern Italian Alps, we compared in-situ air temperature data collected by IoT based sensors below canopies with measurements collected on towers above forest canopies, and with canopy surface temperatures estimated from remote sensing. By integrating these datasets with tree forest inventory data and LiDAR-derived canopy structure metrics, we aimed to quantify the magnitude of canopy thermal buffering and its variation across forest types, canopy structures, diurnal and seasonal cycles and microclimatic conditions. Additionally, we explored the resistance of the buffering effect during drought and heatwave periods.
By yielding more realistic temperate values at which functional processes occur, this research will improve our capacity to study forest ecophysiological responses to climate warming.
How to cite: Andreatta, D., Salvino, L., Dalponte, M., Rodeghiero, M., and Belelli Marchesini, L.: Thermal buffering by forest canopies investigated through remote sensing and IoT sensor networks., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-22183, https://doi.org/10.5194/egusphere-egu26-22183, 2026.
Posters virtual: Tue, 5 May, 14:00–18:00 | vPoster spot 2
EGU26-2734 | ECS | Posters virtual | VPS5
Resource Use Efficiency (RUE) Dynamics of Indian Forests Through an Eco-Hydrogeological Approach Using Machine LearningTue, 05 May, 14:09–14:12 (CEST) vPoster spot 2
Resource Use Efficiency (RUE) serves as a critical indicator of forest ecosystem functionality, reflecting the efficiency of forests in utilizing light, water, and carbon for biomass production. This study investigates the spatiotemporal dynamics of RUE across 14 major Indian forest types from 2014 to 2023 by integrating Light Use Efficiency (LUE), Water Use Efficiency (WUE), and Carbon Use Efficiency (CUE) derived from MODIS satellite products. Using an eco-hydrogeological framework coupled with Random Forest modeling, the study evaluates the influence of climatic, topographic, and hydrological variables on forest productivity. Results reveal considerable spatial heterogeneity and temporal variation in RUE, with the highest efficiencies observed in wet evergreen and semi-evergreen forests and the lowest in dry deciduous and thorn forests. WUE demonstrated substantial variability across forest types and years, particularly impacted by the 2016 drought. CUE was strongly influenced by elevation (R2 = 0.82), and slope emerged as a limiting factor in drier ecosystems. The study highlights that subtropical pine and montane forests exhibit resilience and adaptive efficiency, while arid-zone forests remain vulnerable to climatic stressors. These findings provide actionable insights for sitespecific sustainable forest management and climate resilience planning in India’s diverse forest landscapes.
How to cite: Raj, A. and Kumar, R.: Resource Use Efficiency (RUE) Dynamics of Indian Forests Through an Eco-Hydrogeological Approach Using Machine Learning , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2734, https://doi.org/10.5194/egusphere-egu26-2734, 2026.
EGU26-21910 | Posters virtual | VPS5
Performance and optimisation strategy of a multispectral sensor as part of a newly developed low-cost IoT device for forest monitoring (RemoTrees - beta)Tue, 05 May, 14:12–14:15 (CEST) vPoster spot 2
A new low-cost device based on Internet-of-Things (IoT) communication has been developed within the RemoTrees project to monitor climate-change effects in remote forest ecosystems. One of the key component of this device, referred to as the RemoTrees - beta, is a multispectral chipset composed of four sensors (three AS7265X and one AS7341), providing 26 spectral channels covering the range 410–940 nm. The chipset is equipped with a 1-inch diffuser designed to collect hemispherical solar radiation over incidence angles θ ∈ [−90°, +90°], with an angular response close to the cosine law. Here we present laboratory characterisation and calibration results obtained for 15 replicate RemoTrees - beta units. The spectral performance was highly consistent across devices, with central-wavelength variations below ~2 nm. Full width at half maximum (FWHM) values ranged from 19.17 to 47.93 nm, with standard deviations between 0.32 and 1.74 nm and a maximum relative expanded uncertainty of 0.90%. Because the devices will operate under highly variable illumination conditions (time of day, season, latitude, altitude, cloudiness, and canopy cover), optimisation of integration time (IT) and gain (G) is essential to avoid low digital-number (DN) values and insufficient use of the sensor dynamic range. As commonly applied in field spectrometry, automated IT/G optimisation and scan averaging are recommended to maximise signal-to-noise ratio (SNR) and minimise measurement uncertainty. When IT settings alone are insufficient to reach a satisfactory fraction of the dynamic range (≈65 000 DN; target ≥50%), summing of consecutive readings can be used to effectively increase the integration time while limiting saturation risks under rapidly changing sub-canopy light conditions. Radiometric sensitivity was evaluated by varying G and IT. Under optimised settings, SNR values up to ~5000 were achieved. For AS7265X sensors, gains G > 16 combined with IT optimisation increased SNR by up to ~4×, while for AS7341 gains G > 2 with IT optimisation yielded improvements up to ~5×. Detector nonlinearity contributes an expanded uncertainty of up to ±2.98% (k = 2) if uncorrected, which decreases to ≤±1.24% when nonlinearity correction is applied. The calibration coefficients derived from the tested devices showed moderate inter-device variability, with a maximum variation of approximately 10% for each spectral band. The RemoTrees - beta light sensor demonstrates stable spectral performance, high achievable SNR, and manageable inter-device variability, supporting its suitability for large-scale deployment in forest monitoring networks. Proper optimisation of integration time, gain, and signal averaging is essential to fully exploit the sensor dynamic range and minimise uncertainties under highly variable illumination conditions. Ongoing field deployment will further validate these strategies and refine operational protocols for long-term climate monitoring applications.
How to cite: Mihai, L., Toma, C., Mihalcea, R., Sakowska, K., Vescovo, L., Belelli Marchesini, L., Coppola, V., and Valentini, R.: Performance and optimisation strategy of a multispectral sensor as part of a newly developed low-cost IoT device for forest monitoring (RemoTrees - beta), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21910, https://doi.org/10.5194/egusphere-egu26-21910, 2026.
EGU26-16677 | ECS | Posters virtual | VPS5
Quantifying N₂O Pulses from Millet Croplands: The Role of Drought-Rewetting Cycles Observed via Remote Sensing and CMIP6Tue, 05 May, 14:15–14:18 (CEST) vPoster spot 2
Soil-derived N₂O represents a critical climate feedback in semi-arid agriculture. With a global warming potential (GWP) 298 times greater than CO₂, 6.2 TgN₂O-N is emitted annually from agricultural soils. The atmospheric acceleration with a growth rate >1.0 nmol mol-1 y-1 remains unexplained by fertilizer use alone, suggesting climate change as a critical driver of enhanced emissions, particularly through extreme precipitation and droughts. Rajasthan’s 4.6 million hectares of climate-resilient millet cultivation experience intense droughts and severe monsoon variability. Both these factors lead to drought-rewetting cycles and impact N₂O emissions, which remain unquantified at regional scales.
This study integrates satellite-derived measurements coupled with multi-model ensemble projections to model N₂O emission hotspots in the millet croplands at the district level in this state, which is a major producer of millet in India. Published millet area datasets for spatial distribution and water-filled pore space (WFPS) thresholds (80 - 95%, for optimal denitrification) with soil moisture proxies (NDVI, LST) are integrated to quantify N₂O flux. Standard precipitation index from CMIP6 models (SSP2-4.5, SSP5-8.5) is applied to quantify temporal shifts in wet and dry frequencies.
The results indicated that the denitrification-dominated pathways have dominated during rewetting phases, with N₂O peaks lagging behind soil moisture recovery by 48–72 hours, consistent with the Birch effect. Meta-analytical synthesis suggests rewetting pulses release 5 - 10 times higher N₂O flux than constant moisture conditions. CMIP6 scenarios project 20 - 35% intensification in drought frequency by 2050, driving 15 - 25% increases in cumulative annual N₂O emissions under high-emission scenarios. The regional assessment enables evidence-based fertilizer timing and supports India’s Nationally Determined Contributions (NDCs) by quantifying emissions, thereby paving the way for more effective mitigation strategies.
How to cite: Aarav, P., Burman, P., Phartiyal, G., and Sharma, S.: Quantifying N₂O Pulses from Millet Croplands: The Role of Drought-Rewetting Cycles Observed via Remote Sensing and CMIP6, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16677, https://doi.org/10.5194/egusphere-egu26-16677, 2026.
