@misc{SafonovaGhazaryanStilleretal., author = {Safonova, Anastasiia and Ghazaryan, Gohar and Stiller, Stefan and Main-Knorn, Magdalena and Nendel, Claas and Ryo, Masahiro}, title = {Ten deep learning techniques to address small data problems with remote sensing}, series = {International Journal of Applied Earth Observation and Geoinformation}, volume = {125}, journal = {International Journal of Applied Earth Observation and Geoinformation}, publisher = {Elsevier BV}, issn = {1569-8432}, doi = {10.1016/j.jag.2023.103569}, pages = {17}, abstract = {Researchers and engineers have increasingly used Deep Learning (DL) for a variety of Remote Sensing (RS) tasks. However, data from local observations or via ground truth is often quite limited for training DL models, especially when these models represent key socio-environmental problems, such as the monitoring of extreme, destructive climate events, biodiversity, and sudden changes in ecosystem states. Such cases, also known as small data problems, pose significant methodological challenges. This review summarises these challenges in the RS domain and the possibility of using emerging DL techniques to overcome them. We show that the small data problem is a common challenge across disciplines and scales that results in poor model generalisability and transferability. We then introduce an overview of ten promising DL techniques: transfer learning, self-supervised learning, semi-supervised learning, few-shot learning, zero-shot learning, active learning, weakly supervised learning, multitask learning, process-aware learning, and ensemble learning; we also include a validation technique known as spatial k-fold cross validation. Our particular contribution was to develop a flowchart that helps DL users select which technique to use given by answering a few questions. We hope that our review article facilitate DL applications to tackle societally important environmental problems with limited reference data.}, language = {en} } @misc{MartreDueriGuarinetal., author = {Martre, Pierre and Dueri, Sibylle and Guarin, Jose Rafael and Ewert, Frank and Webber, Heidi and Calderini, Daniel and Molero, Gemma and Reynolds, Matthew and Miralles, Daniel and Garcia, Guillermo and Brown, Hamish and George, Mike and Craigie, Rob and Cohan, Jean-Pierre and Deswarte, Jean-Charles and Slafer, Gustavo A. and Giunta, Francesco and Cammarano, Davide and Ferrise, Roberto and Gaiser, Thomas and Gao, Yujing and Hochman, Zvi and Hoogenboom, Gerrit and Hunt, Leslie A. and Kersebaum, Kurt C. and Nendel, Claas and Padovan, Gloria and Ruane, Alex C. and Srivastava, Amit Kumar and Stella, Tommaso and Supit, Iwan and Thorburn, Peter and Wang, Enli and Wolf, Joost and Zhao, Chuang and Zhao, Zhigan and Asseng, Senthold}, title = {Global needs for nitrogen fertilizer to improve wheat yield under climate change}, series = {Nature Plants}, volume = {10}, journal = {Nature Plants}, number = {7}, publisher = {Springer Science and Business Media LLC}, issn = {2055-0278}, doi = {10.1038/s41477-024-01739-3}, pages = {1081 -- 1090}, language = {en} } @misc{MartreDueriBrownetal., author = {Martre, Pierre and Dueri, Sibylle and Brown, Hamish and Asseng, Senthold and Ewert, Frank and Webber, Heidi and George, Mike and Craigie, Rob and Guarin, Jose Rafael and Pequeno, Diego and Stella, Tommaso and Ahmed, Mukhtar and Alderman, Phillip and Basso, Bruno and Berger, Andres and Bracho Mujica, Gennady and Cammarano, Davide and Chen, Yi and Dumont, Benjamin and Rezaei, Ehsan Eyshi and Fereres, Elias and Ferrise, Roberto and Gaiser, Thomas and Gao, Yujing and Garcia-Vila, Margarita and Gayler, Sebastian and Hochman, Zvi and Hoogenboom, Gerrit and Kersebaum, Kurt C. and Nendel, Claas and Olesen, J{\o}rgen and Padovan, Gloria and Palosuo, Taru and Priesack, Eckart and Pullens, Johannes and Rodr{\´i}guez, Alfredo and R{\"o}tter, Reimund P. and Ruiz Ramos, Margarita and Semenov, Mikhail and Senapati, Nimai and Siebert, Stefan and Srivastava, Amit Kumar and St{\"o}ckle, Claudio and Supit, Iwan and Tao, Fulu and Thorburn, Peter and Wang, Enli and Weber, Tobias and Xiao, Liujun and Zhao, Chuang and Zhao, Jin and Zhao, Zhigan and Zhu, Yan}, title = {Winter wheat experiments to optimize sowing dates and densities in a high-yielding environment in New Zealand: field experiments and AgMIP-Wheat multi-model simulations}, series = {Open Data Journal for Agricultural Research}, volume = {10}, journal = {Open Data Journal for Agricultural Research}, publisher = {Wageningen University and Research}, issn = {2352-6378}, doi = {10.18174/odjar.v10i0.18442}, pages = {14 -- 21}, abstract = {This paper describes the data set that was used to test the accuracy of twenty-nine crop models in simulating the effect of changing sowing dates and sowing densities on wheat productivity for a high-yielding environment in New Zealand. The data includes one winter wheat cultivar (Wakanui) grown during six consecutive years, from 2012-2013 to 2017-2018, at two farms located in Leeston and Wakanui in Canterbury, New Zealand. The simulations were carried out in the framework of the Agricultural Model Intercomparison and Improvement Project for wheat (AgMIP-Wheat). Data include local daily weather data, soil profile characteristics and initial conditions, crop measurements at maturity (grain, stem, chaff and leaf dry weight, ear number and grain number, grain unit dry weight), and at stem elongation and anthesis (total above ground dry biomass, leaf number per stem and leaf area index). Several in-season measurements of the normalized difference vegetation index (NDVI) and the fraction of intercepted photosynthetically active radiation (FIPAR) are also available. The crop model simulations include both daily in-season and end-of-season results from twenty-nine wheat models.}, language = {en} } @misc{CouedelFalconnierAdametal., author = {Cou{\"e}del, Antoine and Falconnier, Gatien N. and Adam, Myriam and Cardinael, R{\´e}mi and Boote, Kenneth and Justes, Eric and Smith, Ward N. and Whitbread, Anthony Michael and Affholder, Fran{\c{c}}ois and Balkovic, Juraj and Basso, Bruno and Bhatia, Arti and Chakrabarti, Bidisha and Chikowo, Regis and Christina, Mathias and Faye, Babacar and Ferchaud, Fabien and Folberth, Christian and Akinseye, Folorunso M. and Gaiser, Thomas and Galdos, Marcelo V. and Gayler, Sebastian and Gorooei, Aram and Grant, Brian and Guibert, Herv{\´e} and Hoogenboom, Gerrit and Kamali, Bahareh and Laub, Moritz and Maureira, Fidel and Mequanint, Fasil and Nendel, Claas and Porter, Cheryl H. and Ripoche, Dominique and Ruane, Alex C. and Rusinamhodzi, Leonard and Sharma, Shikha and Singh, Upendra and Six, Johan and Srivastava, Amit Kumar and Vanlauwe, Bernard and Versini, Antoine and Vianna, Murilo and Webber, Heidi and Weber, Tobias K. D. and Zhang, Congmu and Corbeels, Marc}, title = {Long-term soil organic carbon and crop yield feedbacks differ between 16 soil-crop models in sub-Saharan Africa}, series = {European Journal of Agronomy}, volume = {155}, journal = {European Journal of Agronomy}, publisher = {Elsevier BV}, issn = {1161-0301}, doi = {10.1016/j.eja.2024.127109}, pages = {16}, abstract = {Food insecurity in sub-Saharan Africa is partly due to low staple crop yields, resulting from poor soil fertility and low nutrient inputs. Integrated soil fertility management (ISFM), which includes the combined use of mineral and organic fertilizers, can contribute to increasing yields and sustaining soil organic carbon (SOC) in the long term. Soil-crop simulation models can help assess the performance and trade-offs of a range of crop management practices including ISFM, under current and future climate. Yet, uncertainty in model simulations can be high, resulting from poor model calibration and/or inadequate model structure. Multi-model simulations have been shown to be more robust than those with single models and help understand and reduce modelling uncertainty. In this study, we aim to perform the first multi-model comparison for long-term simulations of crop yield and SOC and their feedbacks in SSA. We evaluated the performance of 16 soil-crop models using data from four long-term maize experiments at sites in SSA with contrasting climates and soils. Each experiment had four treatments: i) no exogenous inputs, ii) addition of mineral nitrogen (N) fertilizer, iii) use of organic amendments, and iv) combined use of mineral and organic inputs. We assessed model performance in two steps: through blind calibration involving a minimum level of experimental data provided to the modeling teams, and subsequently through full calibration, which included a more extensive set of observational data. Model ensemble accuracy was greater with full calibration than blind calibration. Improvement in model accuracy was larger for maize yields (nRMSE 48 vs 18\%) than for topsoil SOC (nRMSE 22 vs 14\%). Model ensemble uncertainty (defined as the coefficient of variation across the 16 models) increased over the duration of the long-term experiments. Uncertainty of SOC simulations increased when organic amendments were used, whilst uncertainty of yield predictions was largest when no inputs were applied. Our study revealed large discrepancies among the models in simulating i) crop-to-soil feedbacks due to uncertainties in simulated carbon coming from roots, and ii) soil-to-crop feedbacks due to large uncertainties in simulated crop N supply from soil organic matter decomposition. These discrepancies were largest when organic amendments were applied. The results highlight the need for long-term experiments in which root and soil N dynamics are monitored. This will provide the corresponding data to improve and calibrate soil-crop models, which will lead to more robust and reliable simulations of SOC and crop productivity, and their interactions.}, language = {en} } @misc{JeltschRoelekeAbdelfattahetal., author = {Jeltsch, Florian and Roeleke, Manuel and Abdelfattah, Ahmed and Arlinghaus, Robert and Berg, Gabriele and Blaum, Niels and De Meester, Luc and Dittmann, Elke and Eccard, Jana Anja and Fournier, Bertrand and Gaedke, Ursula and Gallagher, Cara and Govaert, Lynn and Hauber, Mark and Jeschke, Jonathan M. and Kramer-Schadt, Stephanie and Linst{\"a}dter, Anja and Lucke, Ulrike and Mazza, Valeria and Metzler, Ralf and Nendel, Claas and Radchuk, Viktoriia and Rillig, Matthias C. and Ryo, Masahiro and Scheiter, Katharina and Tiedemann, Ralph and Tietjen, Britta and Voigt, Christian C. and Weithoff, Guntram and Wolinska, Justyna and Zurell, Damaris}, title = {The need for an individual-based global change ecology}, series = {Individual-based ecology}, volume = {1}, journal = {Individual-based ecology}, publisher = {Pensoft Publishers}, address = {Sofia}, issn = {3033-0947}, doi = {10.3897/ibe.1.148200}, pages = {1 -- 18}, abstract = {Biodiversity loss and widespread ecosystem degradation are among the most pressing challenges of our time, requiring urgent action. Yet our understanding of their causes remains limited because prevailing ecological concepts and approaches often overlook the underlying complex interactions of individuals of the same or different species, interacting with each other and with their environment. We propose a paradigm shift in ecological science, moving from simplifying frameworks that use species, population or community averages to an integrative approach that recognizes individual organisms as fundamental agents of ecological change. The urgency of the biodiversity crisis requires such a paradigm shift to advance ecology towards a predictive science by elucidating the causal mechanisms linking individual variation and adaptive behaviour to emergent properties of populations, communities, ecosystems, and ecological interactions with human interventions. Recent advances in computational technologies, sensors, and analytical tools now offer unprecedented opportunities to overcome past challenges and lay the foundation for a truly integrated Individual-Based Global Change Ecology (IBGCE). Unravelling the potential role of individual variability in global change impact analyses will require a systematic combination of empirical, experimental and modelling studies across systems, while taking into account multiple drivers of global change and their interactions. Key priorities include refining theoretical frameworks, developing benchmark models and standardized toolsets, and systematically incorporating individual variation and adaptive behaviour into empirical field work, experiments and predictive models. The emerging synergies between individual-based modelling, big data approaches, and machine learning hold great promise for addressing the inherent complexity of ecosystems. Each step in the development of IBGCE must systematically balance the complexity of the individual perspective with parsimony, computational efficiency, and experimental feasibility. IBGCE aims to unravel and predict the dynamics of biodiversity in the Anthropocene through a comprehensive study of individual organisms, their variability and their interactions. It will provide a critical foundation for considering individual variation and behaviour for future conservation and sustainability management, taking into account individual-to-ecosystem pathways and feedbacks.}, language = {en} } @misc{PalkaNendelWeissetal., author = {Palka, Marlene and Nendel, Claas and Weiß, Lucas and Schiller, Josepha and J{\"a}nicke, Clemens and Gaviria, Juliana Arbel{\´a}ez and Ryo, Masahiro}, title = {Cropping history, agronomic rules, and commodity prices shape crop rotations across Central Europe}, series = {Agricultural systems}, volume = {231}, journal = {Agricultural systems}, publisher = {Elsevier BV}, address = {Amsterdam}, issn = {0308-521X}, doi = {10.1016/j.agsy.2025.104522}, pages = {1 -- 14}, abstract = {Context Crop rotations provide agronomic benefits over monocropping, such as enhanced nitrogen supply, improved weed and pest control, and higher yields. Although the theoretical understanding of optimal rotations has advanced, little is known about their real-world implementation and the factors influencing rotation decisions on large scales. Objective Understanding these factors is key for projecting future cropping patterns, refining agricultural policy, and improving crop models that often oversimplify rotation practices. This study identifies the drivers influencing operational crop rotations across Central Europe and projects future cropping patterns in the region. Methods We analyse over 16 million field-year combinations from Germany, Austria, and the Czech Republic. Using a random forest algorithm, we determine feature importance and apply a novel machine learning approach that incorporates uncertainty in farmers' decision-making to provide a potential outlook on cropping patterns until 2070. Results and Conclusions Historical cropping patterns, agronomic practices, and legume commodity prices significantly shaped crop rotations across the region. Projections indicate a substantial increase in legume cultivation over the coming decades, with implications for nitrogen budgets, dietary transitions, and in-silico upscaling. Significance Rather than optimizing rotations, this study identifies key drivers of operational crop rotations in Central Europe. The findings provide the basis for large-scale simulations that represent cropping patterns more realistically. To the best of our knowledge, the data set compiled here is the most extensive yet analysed in the context of operational crop rotation management.}, language = {en} } @misc{RazaRyoGhazaryanetal., author = {Raza, Ahsan and Ryo, Masahiro and Ghazaryan, Gohar and Baatz, Roland and Main-Knorn, Magdalena and Inforsato, Leonardo and Nendel, Claas}, title = {Predicting regional-scale groundwater levels at high spatial resolution using spatial random forest models}, series = {International journal of applied earth observation and geoinformation}, volume = {144}, journal = {International journal of applied earth observation and geoinformation}, publisher = {Elsevier BV}, address = {Amsterdam}, issn = {1569-8432}, doi = {10.1016/j.jag.2025.104918}, pages = {1 -- 19}, abstract = {Groundwater is an important resource for sustainable crop growth and agricultural productivity, and groundwater level (GWL) fluctuations directly influence agroecosystems. However, GWL data for most regions is unavailable because of spatial scarcity and discontinuous groundwater observations. In recent studies, machine learning techniques have proven to be a useful tool for producing GWL estimates, but these techniques may not necessarily take into account distances and observations from the nearest points to unmonitored sites as covariates. In this study, we explored the potential of these additional spatial covariates by implementing Random Forest for Spatial Interpolation (RFSI) to generate high-resolution (1 km²) GWL estimates on a monthly scale. To evaluate the effectiveness of the RFSI model, we performed a thorough comparison of the RFSI model with a conventional Random Forest (RF) model, a Random Forest for spatial data (RFsp) and a Support Vector Machine (SVM). The framework was implemented using GWL observation data for the period 2001 to 2022 in Brandenburg, Germany. The RFSI model exhibited the highest predictive accuracy for GWL predictions at testing points, achieving an R² value of 0.86, surpassing the performance of RF, RFsp and SVM with R² values of 0.75, 0.72 and 0.67, respectively. Incorporating nearby-well information substantially improved 1 km GWL predictions: RFSI achieved the lowest RMSE (3.77 m), as compared to RF (4.85 m), RFsp (5.04 m), and SVM (5.59 m). Notably, among the models tested, RF, RFsp, and SVM showed notable inaccuracies, substantially underestimating groundwater levels (GWL) in deeper wells (>10 m) and overestimating levels in shallow wells (1-10 m). In contrast, RFSI performed better, with smaller overestimations (up to ~1.1 m) and less severe underestimations (up to ~-3.9 m). Furthermore, the RFSI model demonstrated significantly higher computational efficiency compared to RFsp and SVM, particularly for large training datasets and high-resolution prediction mapping.}, language = {en} }