Research and dissemination
Background activities
Background activities
Bridging the Trans Scalar Disconnect in Ecohydrology
Current Earth observation workflows frequently rely on optical indices. We consider this a major methodological weakness because these indices are fundamentally reactive. They quantify structural degradation and chlorophyll depletion only after irreversible physiological failure has already occurred. Furthermore, global monitoring faces a severe spatial disconnect. The mechanistic drivers of plant mortality operate at the tissue scale, whereas satellite monitoring relies on coarse macroscopic footprints. To overcome this extreme nonlinearity, OpenSkyLab introduces the TENSION framework. We utilise physics informed data assimilation to bridge this spatial gap and invert high resolution drone data into continuous spatial predictions of internal hydraulic tension.
Overcoming the Pre Dawn Equilibrium Fallacy
A persistent bad practice in global ecohydrological modelling is the universal assumption of pre dawn equilibrium between leaf and soil water potential. During prolonged compound drought and heatwaves, nocturnal transpiration surpasses internal hydraulic recharge, which renders standard equilibrium assumptions invalid. By developing rigorous in situ monitoring protocols, we explicitly quantify this nocturnal hydraulic capacitance. Correcting this fallacy is a crucial step for accurately initialising morning water states in our models and represents a vital lesson learned for the entire remote sensing community.
Functional Isolation and the Digital Twin
To identify irreversible ecosystem tipping points before structural degradation occurs, we must isolate pure physiological signals. Our approach relies on separating dynamic photochemical yield from structural canopy distortions and soil evaporation. This rigorous standardisation provides a pre visual spatial proxy for internal xylem tension and vulnerability thresholds. We then assimilate these multi domain datasets into a coupled mechanistic model. This architecture creates a predictive digital twin that is capable of robust spatial inversion without relying on simplistic empirical scaling.
Macroscopic Early Warning Systems
The overarching objective of the TENSION project is to validate the spatial transferability of local hydraulic vulnerability to regional scales. By fusing drone derived functional metrics with macroscopic satellite signals, we aim to detect high frequency anomalies in canopy water content. This integration identifies regional vulnerability hotspots prior to any detectable optical decline. Ultimately, this approach delivers a physically validated early warning framework for ecosystem resilience and definitively shifts environmental monitoring from reactive observation to presymptomatic prediction.
Behind the OpenSkyLab Idea
Drones deliver unmatched detail for monitoring our environment, but their potential is often limited by fragmented workflows, missing standards and poor reproducibility. OpenSkyLab was created to change this paradigm and bring the remote sensing community together. Each dataset on our portal is paired with a comprehensive metadata package covering flight parameters, calibration, environmental conditions and processing provenance. This makes data transparent, traceable and reusable across sites, times and sensor types.
The Challenge of Uncertainty
Environmental remote sensing has advanced rapidly with the rise of drones, high resolution sensors and expanding satellite archives. These tools have opened new possibilities for monitoring biodiversity, detecting vegetation stress and mapping surface temperature. Yet persistent uncertainty remains a major challenge. Variable weather conditions, sensor limitations and divergent workflows can degrade data quality, hinder comparability and weaken reproducibility, especially in structurally complex landscapes such as heterogeneous farmland and mixed forests.
Rigorous Metadata and Corrections
Experience from recent studies highlights that reliable monitoring depends on systematic metadata collection. Flight parameters, sensor settings and weather conditions during acquisition must be meticulously documented to ensure results are interpretable and repeatable. Robust correction procedures are equally essential. Radiometric and atmospheric calibration, geometric alignment and sensor harmonisation underpin any comparison across dates, sites or platforms. Without such steps, vegetation indices or surface temperature estimates can be highly unstable and misleading.
Workflow Tailoring and Interdisciplinary Collaboration
Another consistent finding is that there is no universal workflow. Pipelines must be tailored to the environment and the variable of interest, and transparency about assumptions improves transferability. Larger and better stratified samples reduce variance more effectively than post hoc modelling fixes. Furthermore, multisensor integration strengthens robustness only when geometric and radiometric consistency are maintained. Ultimately, progress depends on close cooperation between ecology and computer science to design methods that are transferable, scalable and ecologically meaningful.
Building an Interoperable Network
These lessons emphasise that advancing environmental monitoring will rely less on ever new sensors and more on smarter deployment strategies, careful correction workflows and standardised data handling. Drone observations can only fulfil their potential when accompanied by rigorous validation and a commitment to reproducibility. Through an interactive map, open downloads and international metadata standards, OpenSkyLab turns scattered one off surveys into a shared and interoperable network that provides a solid foundation for long term environmental monitoring.
Moving Beyond
Uncalibrated Imagery
Thermal and moisture conditions are key drivers of ecosystem processes, and obtaining reliable high resolution data remains a major challenge. At first glance, drone thermal imagery appears to offer an easy solution for diverse applications. However, we often highlight this assumption as a common bad practice. Accurately estimating land surface temperature and moisture from such imagery is a significant hurdle that requires strict methodological discipline. Thermal data are strongly influenced by variable vegetation emissivity, sensor viewing geometry, distance to the target and changing atmospheric conditions. Without appropriate physical corrections and detailed metadata records, these factors introduce strong biases that lead to substantial overestimations or underestimations. Simply flying a thermal camera over a field without ground calibration generates uninterpretable pictures rather than scientific data.
Developing a Robust Correction Framework
To address these critical gaps, we develope and share a robust correction framework. We integrate drone thermal imagery with ground meteorological observations and vegetation indices to enforce physical consistency. Our workflow incorporates relative humidity, wind speed and derived emissivity to reduce systematic errors. By applying atmospheric correction models to compensate for transmittance loss and using ground reference stations as temperature benchmarks, we significantly enhance the accuracy of surface temperature estimates. This methodological rigour allows us to confidently derive soil and canopy moisture proxies.
Scalability and Educational Impact
We view this not just as a research tool but as a best practice standard for the community. The resulting workflow provides a scalable method that links drone observations directly to ground conditions across time series. It transforms fragmented mapping efforts into reliable insights for biodiversity monitoring, precision agriculture and ecosystem management. Furthermore, we actively incorporate these lessons learned into our educational programmes to teach the next generation of researchers how to avoid common thermal mapping pitfalls.
Addressing Remaining Challenges
Despite these advances, several challenges remain and we openly acknowledge them to maintain scientific transparency. Thermal drift in lightweight sensors, rapid atmospheric fluctuations during flight and mixed pixel effects in heterogeneous canopies can still introduce uncertainty. Our ongoing work focuses on automating calibration routines, integrating radiative transfer models and testing the framework across diverse climatic regions to assess robustness and transferability. Ultimately, integrating drone thermal data with in situ measurements and rigorous metadata is the only pathway toward robust land and atmosphere monitoring. By bridging the gap between drone scale measurements and ground observations, our framework offers a transparent approach to surface temperature and moisture estimation.
From Forest Health to Infrastructure Safety
Whether addressing climate induced ecological crises or managing critical linear infrastructure, the fundamental challenge in environmental remote sensing remains the same. We need to transform raw and high-resolution data into actionable and reproducible science at the level of individual trees. We developed and validated robust drone-based workflows that replace isolated case studies with scalable and harmonised processing pipelines.
Methodological Synergy: Capturing Symptoms and Causes
Our core approach integrates high-resolution time series of multispectral, thermal and RGB imagery alongside precise structural models derived from photogrammetry and lidar. We move beyond simple mapping by focusing on a mechanistic understanding of vegetation dynamics. Multispectral data provide crucial information about leaf pigment composition and canopy structure, effectively revealing the onset and type of stress, which we consider the symptoms. Concurrently, thermal imagery records emitted longwave radiation. When plants close their stomata to reduce transpiration under water stress, leaf temperatures rise. Thermal data therefore provide a direct physical indicator of stress intensity and reveal the causes. By coupling these multi-sensor approaches with machine and deep learning algorithms, we create highly accurate spatial databases of individual trees and shrubs.
Applied Workflows in Action
We systematically apply these rigorous methodologies to two pressing environmental domains. First, prolonged droughts and rising temperatures across Central Europe have triggered unprecedented bark beetle infestations. By tracking canopy health over time, our algorithms detect pre-visual and early-warning signals of water stress and pest attack. This allows forest managers to implement timely sanitation and sustainable interventions.
Second, managing vegetation along remote linear transport corridors is vital for mitigating safety hazards and for monitoring invasive species. Our semi-automated workflows deliver spatially explicit risk maps, optimise maintenance schedules and reduce long-term economic and ecological costs.
Best Practices and the Open Data Standard
A central pillar of OpenSkyLab is our commitment to methodological transparency. We actively identify gaps and bad practices in current remote sensing applications and advocate for robust data provenance. Transparent metadata, open data policies and field-validated models ensure that our workflows are reproducible and transferable. By sharing lessons learned, we empower foresters, transport authorities and researchers to adopt long-term and policy-relevant monitoring frameworks.
