Publications
NIBIOs employees contribute to several hundred scientific articles and research reports every year. You can browse or search in our collection which contains references and links to these publications as well as other research and dissemination activities. The collection is continously updated with new and historical material.
2026
Abstract
This paper presents a social benefit-cost analysis (BCA) of Precision Agriculture (PA)and Enhanced Efficiency Fertilizers (EEF) for reducing nitrate pollution in the Zelivka catchment, the drinking-water source for 1.5 million people in Prague and surrounding regions. The analysis combines field-level data (50,309 ha, 729 farms), meta-analytical biophysical parameters, a dynamic heterogeneous-agent adoption model, and cross-validation against an independent SWAT+ hydrological simulation. Technology adoption generates a mean Net Present Value of EUR 291 million (95% CI: EUR 94 775 million) over fifty years, with a deterministic benefit-cost ratio of 19.8. Even with health benefits set to zero, the NPV remains positive at EUR 44 million, indicating that private benefits alone cover adoption costs. BCA and the SWAT+ approaches imply comparable rates of reservoir-concentration decline (≈0.15 mg/L/yr in SWAT+ over its lag-free window versus a ≈0.13 mg/L/yr peak in the BCA once the lag has elapsed). At-source nitrogen prevention via agricultural technology adoption offers a cost-effective complement to structural catchment measures and end-of-pipe water treatment.
Authors
Helle Ross GobakkenAbstract
Support from Veiflyt project financed by Skogtiltaksfondet
Authors
Lokesh P Padhye Melanie Kah Erin M Leitao Karl Bowles Paul Nathanail Ian T Cousins Romain Figuière Bradley O Clarke Jordan M Partington Wejdan Alghamdi Elsie M Sunderland Bridger J Ruyle Satoshi Endo Trevor N Brown Zhengyang Wang Size Zheng Joseph J Pignatello Sanne J Smith Marcel Riegel Hans Peter Heinrich Arp Jens Blotevogel Robert J Giraud Anthony K Rappé Erlend Grenager Sørmo Gerard Cornelissen Marc A Deshusses Igor V Novosselov P Lee Ferguson Brian R Pinkard Timothy J Strathmann Kapish Gobindlal Jonathan Sperry Elisabeth Cuervo Lumbaque Nick Duinslaeger Jelena Radjenovic James HattonAbstract
Per- and polyfluoroalkyl substances (PFAS) represent a large—and structurally diverse—group of contaminants that have become ubiquitous in our environment. PFAS are all extremely persistent while some are also bioaccumulative, mobile and/or toxic, which gives rise to significant environmental and health concerns. Despite more than a decade of intensive research, the management of PFAS is still associated with considerable challenges. It is evident that a holistic approach is required to address the challenging global problem of PFAS. This roadmap features expert perspectives from world-renowned leading researchers and practitioners on how best to manage PFAS. The 15 topics cover different facets of the complex PFAS issue, providing a multidisciplinary and multisectoral overview. For each topic, we reflect on the current status of knowledge and offer recommendations on science and technology advances that will help meet current and future challenges. Taken together, the 15 topics cover the entire life cycle of PFAS—from their sources to their destruction. Important themes such as monitoring and analysis, understanding and predicting fate, source controls (regulation and replacement), and existing and emerging strategies for remediation (capture and destroy) are highlighted throughout the roadmap. Overall, there are many recent scientific and technological advancements that show promise for the management of PFAS. However, it is also clear that there is no ‘silver bullet’ and multifaceted solutions will be needed. Long-term success hinges on sustained collaboration among researchers, policymakers, industries, and communities, which we hope this roadmap will help to catalyze.
Authors
A. Terskaia R. Oucheikh Junxiang Peng J. Oliveira D. Parsons J. Morel A.M. Abdi Z. Cai S. Thapa N. Boke-Olen P. Olsson H.G. Smith J. Ekroos L. EklundhAbstract
Grassland management practices are critical both for maintaining resilient food production and for preserving biodiversity, yet spatially explicit information on them is often limited. Recent advances in freely available, remotely sensed land surface phenology and productivity datasets raise the question of whether phenological metrics provide additional, biologically interpretable information for characterising and mapping grassland management beyond conventional spectral features. In this study, we evaluated the information content and added value of freely available, ready-to-use Sentinel-2–derived High-Resolution Vegetation Phenology and Productivity (HR-VPP) data for distinguishing broad grassland management types (conventional leys, organic leys, and semi-natural grasslands) across multiple spatial scales. We analyzed 1455 grasslands across the intensively farmed Skåne province (southern Sweden) over 2018–2022 at provincial and local scales. We further tested whether phenological variables improve grassland management classification when combined with multi-temporal Sentinel-2 spectral data using state-of-the-art machine learning (ML) methods, and applied a conformal prediction framework to manage classification uncertainty. Relationships between HR-VPP metrics and field-measured biomass were additionally assessed using independent field data from northern Sweden. Field-level analyses showed weak (R2 = 0.28) to moderate (R2 = 0.47) relationships between HR-VPP productivity metrics and measured biomass. HR-VPP metrics captured statistically significant but subtle differences among management types, with strong overlap between categories, particularly at the provincial scale. Differences were more pronounced at local scales. Consequently, HR-VPP variables alone had limited predictive power and ML model accuracy was low (0.53). However, combining HR-VPP with Sentinel-2 spectral time-series data substantially improved classification performance, achieving an accuracy of 0.70. Feature importance analysis indicated that spectral indices dominated classification, with limited direct contribution from phenological parameters. These findings highlight both the potential and current limitations of continental-scale phenology and productivity products for grassland management monitoring and point to the need for improved models and complementary data sources to better capture management practices.
Authors
Telmo Jorge Carneiro Amado Jackson Ernani Fiorin William Ramos da Silva Thiago Inagaki Ulfried Arns Tiago Wyzykowski Felipe José Cury Fracetto Ademir de Oliveira FerreiraAbstract
Diversified no-tillage crop successions are key strategies for improving soil health, productivity, and sustainability in tropical environments. However, adoption of diversified cropping systems remains limited due to economic, cultural, and technical constraints. This study aimed to evaluate how long-term diversified crop rotations, differing in legume inclusion and mineral phosphorus (P) and potassium (K) fertilization, influence system nitrogen (N), P, and K budgets, biomass production, and crop productivity in mature tropical no-tillage systems (NTS). Using a 10-year field experiment established on a tropical Ferralsol in southern Brazil, we evaluated six long-term crop rotations that differed in crop diversification, inclusion of Fabaceae cover crops, and phosphorus and potassium mineral fertilization. Total system N, P, and K budgets were quantified by integrating nutrient inputs, crop uptake, grain nutrient export, biomass nutrient cycling, and net soil nutrient balances. Intensive legume-based rotations increased crop productivity compared with low-diversity, non-legume rotations (specifically, the two-crop black oat/common bean succession). Vetch-based rotations under optimal or suboptimal mineral fertilization, as well as unfertilized vetch rotations, produced the highest cumulative grain yields (∼52,000 kg ha -1 ), representing a 38% increase (a 1.38-fold increase) over the black oat/common bean rotation (∼38,000 kg ha -1 ). This response was associated with a 40% increase in aboveground biomass and grain yield when maize followed hairy vetch rather than black oat. Legume-intensive rotations under optimal mineral fertilization or without fertilization maintained a strongly positive N budget (+559 kg ha -1 ), whereas the low-diversity black oat/common bean rotation showed a net N deficit. Conversely, all unfertilized systems, including the highly productive unfertilized hairy vetch-based rotations, exhibited strongly negative P and K budgets, indicating a progressive risk of soil fertility depletion. Only suboptimal mineral-fertilized vetch rotations partially offset these nutrient deficits in highly crop-productive systems. The study highlights the role of legume-based crop rotations as a key strategy to sustain a positive N budget, thereby supporting nutrient cycling and crop productivity under tropical NTS. Maintaining nutrient balance remained dependent on coupling biological N fixation with balanced mineral P and K fertilization to offset nutrient export deficits in high-yield systems and to reduce the risk of long-term soil fertility depletion.
Authors
Andrea Ponzecchi Albin Lobo Claudia Cocozza Paolo Cherubini Matthias Saurer Erik Larnøy Lisbeth G. ThygesenAbstract
Broadleaf afforestation increasingly requires planting material that combines valuable wood properties with resilience to climatic stress. High heartwood content is a desirable trait in pedunculate oak (Quercus robur), but it is unclear whether selecting for it could be a trade-off with resistance and resilience to drought. We addressed this question in a 24 year old progeny trial in southern Denmark where trees were ranked by heartwood production and the 2018 drought was used as a natural stress event. Trees with contrasting heartwood content were assessed using multiple indicators of drought response, including annual ring width, vessel anatomy, wood density, and dual stable isotopes (δ¹³C and δ¹⁸O) measured separately in earlywood and latewood over a five-year period. The 2018 summer drought was of moderate intensity at the trial site, with SPEI3 approximately −1.5, but it was clearly reflected in isotopic signals, indicating that trees perceived the water deficit. In contrast, radial growth did not decline, and ring widths remained above average in 2018. Across all metrics, trees with lower and higher heartwood content showed broadly similar physiological and wood-structural responses. Within the environmental conditions and drought intensity observed in this study, our results provide no evidence for a strong trade-off between heartwood content and response to a drought of moderate severity. They therefore support, with appropriate caution, the continued consideration of heartwood content as a selection trait in breeding and afforestation programs in comparable environments.
Abstract
Background Faba bean cultivation in northern Europe is challenged by high variability in yield performance due to strong genotype × environment interaction (GEI). Appropriate GEI modeling and interpretation is essential for both cultivar recommendation and breeding for Nordic conditions. Methods Grain yield data from fifteen spring faba bean cultivars evaluated in eleven Norwegian environments during four growing seasons (2022–2025) were analyzed using the Additive Main Effects and Multiplicative Interaction (AMMI) model. Climatic drivers of GEI were investigated using factorial regression, while genome-wide SNP markers were used to assess the relationship between molecular diversity and adaptive differentiation. Results Environment, genotype, and GEI significantly influenced grain yield. The AMMI model with three GEI principal components (AMMI3) represented the biologically meaningful GEI signal and explained 83.4% of the GEI variation. The predictive accuracy of this model closely approached that of BLUP and GBLUP. Factorial regression showed that maximum and mean September temperatures explained 38.5% of the GEI variation, indicating that late-season thermal conditions are the principal environmental drivers of cultivar adaptation. We identified three top-yielding genotypes (Ketu, Birgit, and Vire) that can be object of site-specific recommendation. Two sub-regions emerged as possibly distinct breeding targets, namely, a major one, and a smaller one featuring cooler late-cycle temperatures and specific adaptation of very early material. A significant Mantel correlation between genomic and adaptive dissimilarity (r = 0.40, P = 0.029) suggested that differences in environmental adaptation have a measurable genetic basis. Conclusions This study identified elite cultivars for recommendation and generated crucial information for future crop breeding and variety testing in Norway. Breeding for wide adaptation across climatically diversified environments is justified by the modest size of the smaller sub-region and its expected decreasing importance due to climate change. The relationship between genomic and adaptive dissimilarity, if confirmed for a larger genotype set, could be exploited for a preliminary screening of regional breeding material and novel plant introductions.
Abstract
As forests face increasing anthropogenic pressure, the robust, large-scale remote sensing of key biophysical parameters, such as forest height and above-ground biomass density, has become essential to guide conservation efforts and climatological analyses. While deep learning (DL) applied to interferometric synthetic aperture radar (InSAR) data has achieved state-of-the-art performance, its operational potential for long-term forest monitoring remains limited by a lack of uncertainty reporting, which is essential for error propagation, and for evaluating prediction reliability and temporal stability. This study introduces a novel Bayesian DL framework for the uncertainty-aware estimation of forest height from single-pass TanDEM-X InSAR data, marking its first application to synthetic aperture radar interferometry. The approach is validated through a case study in Norway, using national airborne laser scanning data to assess the calibration of the self-reported uncertainties, and to examine the effects of tree species variability and temporal mismatches in training pairs on model generalization. To capture total predictive uncertainty, the framework explicitly models the aleatoric component during estimation while integrating epistemic uncertainty through the comparison of different intra- and interbasin approximation strategies. In addition, the robustness of the model is evaluated under out-of-distribution (OOD) conditions (i.e., data domains absent from the training set), reflecting the challenges encountered in operational remote sensing. The results demonstrate robust generalization performance and the generation of well-calibrated uncertainty maps under in-distribution conditions, while highlighting the critical necessity of complementing data-driven models with an ad hoc OOD detector to reliably manage OOD scenarios. The resulting framework paves the way for real-world prediction of uncertainty-aware forest canopy height products from InSAR data.
Authors
Cordula I.H. Wittekind Felix Witing Sydney E. White Christoph Schürz Marie Anne Eurie Forio Moritz Shore Csilla Farkas Martin Futter Dennis Collentine Lorenzo Sanguanini Ronald A. Kolcsar Brigitta Szabo Piroska Kassai Martin Volk Michael StrauchAbstract
This study presents ParetoPick-R, an R Shiny application for post-processing multi-objective optimisation (MOO) results in environmental modelling. Although MOO is widely used in environmental management, its multi-dimensional and abstract outputs are often hard to interpret and apply directly. ParetoPick-R addresses this challenge by enhancing the interpretability and usability of MOO results for decision support. The tool integrates three key strategies. First, interactive visualisation techniques enable users to explore synergies and trade-offs among objectives and link objective and decision spaces. Second, a clustering-based data reduction approach condenses large Pareto sets into representative subsets while preserving essential trade-off structures. Third, an Analytical Hierarchy Process facilitates the systematic incorporation of stakeholder preferences across objectives and decision variables. This paper describes the design and implementation of ParetoPick-R and demonstrates its functionality and procedural application through two distinct case studies.
Authors
Adrian Straker Paul Magdon Marco Zullich Maximilian Freudenberg Christoph Kleinn Johannes Breidenbach Stefano Puliti Nils NölkeAbstract
Deep learning-based tree species classification models applied to terrestrial laser scanning data achieve high accuracies, yet their classification decisions remain poorly understood. Addressing this gap, we present a novel explanation framework that systematically links salient regions of Finer-CAM saliency maps—which highlight image features contributing to the classification of a target species while suppressing features shared with similar species—to structural tree features in 2D side-view images of terrestrial laser scanning (TLS) point clouds. We applied this framework using YOLOv8-based tree species classification models trained on a subset of the FOR-species20K benchmark data set. Faithfulness analysis confirmed that Finer-CAM reliably identifies the most discriminative image regions for model predictions, validating its suitability for enhancing model interpretability. Analysis of 630 saliency maps indicate that the models primarily rely on image regions associated with tree crowns across most species, particularly for Silver Birch, European Beech, English Oak, and Norway Spruce, while stem regions contribute more strongly to the differentiation of European Ash, Scots Pine, and Douglas-fir. Image perturbation experiments further demonstrated that the visibility of tree structure—such as branching patterns—in the 2D side-view images enhances classification performance, indicating YOLOv8’s capabilities to leverage detailed structural information encoded in high-density TLS point clouds. Using a cross-validation approach our trained models achieved a mean overall accuracy of 96% (SD = 0.24%) on a test data set unseen during training. Beyond assessing the contribution of image regions associated with structural tree features to tree species differentiation, our methodological framework can facilitate a better understanding of model limitations, the detection of data set biases and shortcut learning, and contributes in building confidence in deep learning-based tree species classifiers.