Search

16 pages, 6 figures, 3 tables, supplementary data https://doi.org/10.1016/j.pocean.2021.102659.-- Code and data availability: The experimental protocol in this paper has no code associated with it. Forcing data from CMIP5 used for the protocol, and the FishMIP model outputs presented in this paper are available on the ISIMIP servers (https://www.isimip.org/), Climate change is warming the ocean and impacting lower trophic level (LTL) organisms. Marine ecosystem models can provide estimates of how these changes will propagate to larger animals and impact societal services such as fisheries, but at present these estimates vary widely. A better understanding of what drives this inter-model variation will improve our ability to project fisheries and other ecosystem services into the future, while also helping to identify uncertainties in process understanding. Here, we explore the mechanisms that underlie the diversity of responses to changes in temperature and LTLs in eight global marine ecosystem models from the Fisheries and Marine Ecosystem Model Intercomparison Project (FishMIP). Temperature and LTL impacts on total consumer biomass and ecosystem structure (defined as the relative change of small and large organism biomass) were isolated using a comparative experimental protocol. Total model biomass varied between −35% to +3% in response to warming, and -17% to +15% in response to LTL changes. There was little consensus about the spatial redistribution of biomass or changes in the balance between small and large organisms (ecosystem structure) in response to warming, an LTL impacts on total consumer biomass varied depending on the choice of LTL forcing terms. Overall, climate change impacts on consumer biomass and ecosystem structure are well approximated by the sum of temperature and LTL impacts, indicating an absence of nonlinear interaction between the models’ drivers. Our results highlight a lack of theoretical clarity about how to represent fundamental ecological mechanisms, most importantly how temperature impacts scale from individual to ecosystem level, and the need to better understand the two-way coupling between LTL organisms and consumers. We finish by identifying future research needs to strengthen global marine ecosystem modelling and improve projections of climate change impacts, JDE was funded by Australian Research Council Discovery Projects DP150102656 and DP190102293. MC, JS, NB and OM received financial support by the European Union’s Horizon 2020 research and innovation programme under grant agreement No 817578 (Triatlas project). CH received funding from the Open Philanthropy Project. NB and OM also acknowledge the support of the French ANR project CIGOEF (grant ANR-17-CE32-0008-01). DPT acknowledges funding from the ISI-MIP project to support a workshop on this topic, and the Jarislowsky Foundation. EDG received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 682602, BIGSEA). RFH was funded by the Spanish Ministry of Science, Innovation and Universities through the Acciones de Programación Conjunta Internacional (PCIN-2017-115). MC acknowledges the ‘Severo Ochoa Centre of Excellence’ accreditation (CEX2019-000928-S) to the Institute of Marine Science (ICM-CSIC). TDE acknowledges funding from the ISIMIP project to support a workshop on this topic and the Fisheries and Oceans Canada Atlantic Fisheries Fund. All authors declare no conflict of interest with respect to this study. JAFS received funding from the European Union’s Horizon 2020 FutureMARES project (#869300).

The ocean's “biological pump” significantly modulates atmospheric carbon dioxide levels. However, the complexity and variability of processes involved introduces uncertainty in interpretation of transient observations and future climate projections. Much research has focused on “parametric uncertainty,” particularly determining the exponent(s) of a power‐law relationship of sinking particle flux with depth. Varying this relationship's functional form introduces additional “structural uncertainty.” We use an ocean biogeochemistry model substituting six alternative remineralization profiles fit to a reference power‐law curve, to systematically characterize structural uncertainty, which, in atmospheric pCO2 terms, is roughly 50% of parametric uncertainty associated with varying the power‐law exponent within its plausible global range, and similar to uncertainty associated with regional variation in power‐law exponents. The substantial contribution of structural uncertainty to total uncertainty highlights the need to improve characterization of biological pump processes, and compare the performance of different profiles within Earth System Models to obtain better constrained climate projections., Key Points Six alternative flux profiles fit to a Martin Curve yield large differences in atmospheric carbonStructural uncertainty comprises one‐third of total uncertainty in the ocean's biological pumpVarying particle attenuation with depth may account for half of the biological pump's overall carbon drawdown

Karst aquifers are characterized by high-conductivity conduits embedded in a low-conductivity fractured matrix, resulting in extreme heterogeneity and variable groundwater flow behavior. The conduit network controls groundwater flow, but is often unmapped, making it difficult to apply numerical models to predict system behavior. This paper presents a multi-model ensemble method to represent structural and conceptual uncertainty inherent in simulation of systems with limited spatial information, and to guide data collection. The study tests the new method by applying it to a well-mapped, geologically complex long-term study site: the Gottesacker alpine karst system (Austria/Germany). The ensemble generation process, linking existing tools, consists of three steps: creating 3D geologic models using GemPy (a Python package), generating multiple conduit networks constrained by the geology using the Stochastic Karst Simulator (a MATLAB script), and, finally, running multiple flow simulations through each network using the Storm Water Management Model (C-based software) to reject nonbehavioral models based on the fit of the simulated spring discharge to the observed discharge. This approach captures a diversity of plausible system configurations and behaviors using minimal initial data. The ensemble can then be used to explore the importance of hydraulic flow parameters, and to guide additional data collection. For the ensemble generated in this study, the network structure was more determinant of flow behavior than the hydraulic parameters, but multiple different structures yielded similar fits to the observed flow behavior. This suggests that while modeling multiple network structures is important, additional types of data are needed to discriminate between networks., National Science Foundation Graduate Research Fellowship Program

Sources of uncertainty in a marine biogeochemical model include input from physical processes and the choice of functional forms representing the strength and dependencies of biogeochemical processes. This study explores characteristic signatures from these uncertainties by generating ensembles from perturbing the biogeochemistry equations and perturbing physical input using a 1-D intermediately-complex model run at five oceanographic stations. Perturbed biogeochemistry ensemble (PBE) produces larger spreads than perturbed physics ensemble (PPE), and distinctly different ensemble variations. Fractions of nitrogen in phytoplankton pool from observations show a larger variability than in any single model-ensemble member, but the PBE spread generally captures this variability, whereas the PPE spread does not. The results show that the PBE method\ud gives a more realistic representation of uncertainty than PPE in our 1D-model setup. Our method needs to be tested in more complex models in order to understand its significance on larger scales.

In this introductory paper to the special issue on crop model prediction uncertainty, we present and compare the methodological choices in the studies included in this issue, and highlight some remaining challenges. As a common framework for all studies, we define prediction uncertainty as the distribution of prediction error, which can be written as the sum of a bias plus a predictor uncertainty term that represents the random variation due to uncertainty in model structure, model parameters or model inputs. Several themes recur in many of the studies: Use of multi-model ensembles (MMEs) to quantify model structural uncertainty; Emphasis on uncertainty in those inputs related to prediction of regional results or climate change impact assessment; Simultaneous consideration of multiple sources of uncertainty; Emphasis on exploring the variability of uncertainty over space or time; Use of sensitivity analysis techniques to disaggregate the separate contributions to prediction uncertainty. Relatively new approaches include the estimation of both the bias and predictor uncertainty terms of prediction error, the construction of MMEs specifically designed to explore the uncertainty in model structure, the use of emulators for sensitivity analysis and the exploration of ways to reduce prediction uncertainty other than through model improvement. Major remaining challenges are standardization of approaches to quantifying uncertainty in model structure, parameters and inputs, going beyond studies of specific sources of uncertainty to estimation of overall prediction uncertainty, comparing and combining validation and uncertainty studies, and evaluation of uncertainty estimates. Looking forward, we suggest that assessment of prediction uncertainty should be a standard part of any modelling project. The studies here will contribute toward that goal.

Integrated assessment (IA) modelling can be an effective tool to gain insight into the dynamics of coupled earth system (land use, climate etc.) and socio-economic components. Quantifying and communicating uncertainties is a challenge of any scientific assessment, but is here magnified by the complex and boundary-crossing nature of IA models. Understanding the dynamics of coupled earth and socio-economic systems require data and methods from multiple disciplines, each with its own perspective on epistemological uncertainties (parametric and structural uncertainties), and its own protocols for assessing uncertainty. During the Paris Agreement, the lack of uncertainty analyses (UA) in IAs was risen ( Rogelj et al. 2017 ) and calls for close collaboration of scientists coming from different fields. In this study, we review how uncertainties are tackled in a range of science disciplines that are related to global change including climate, hydrology, energy and land use, and which contribute to IA modelling. We conducted a meta-analysis to identify the contributing disciplines, and review which type of uncertainties are assessed. We then describe sources of uncertainty (e.g. parameter values, model structure), and present opportunities for improved assessment and communication of uncertainties in IA modelling. We show in our meta-analysis that parametric uncertainty is the uncertainty analysis that has been applied the most, while structural uncertainty is less commonly applied, with the exception of the energy scientific discipline. We finish our study with key recommendations to improve uncertainty analysis such as including risk analysis. By embracing uncertainties, resilient and effective solutions for climate change mitigation and adaptation could be better communicated, identified and implemented.

This study presents a novel expansion of the clay-fraction (CF)/resistivity clustering method aiming at developing realizations of subsurface structures based on multiple point statistics (MPS). The CF-resistivity clustering method is used to define a data driven training image (TI) for MPS simulations. By combining this TI with uncertainty estimates obtained from correlation between the resistivity models and the unique categories in the TI, subsurface realizations are generated honoring all geophysical and lithological data including any equivalent resistivity layer sequences. The generated subsurface realizations were calibrated in a steady state groundwater model. Forecasts of well catchment zones were derived based on two wells located in areas with different levels of structural uncertainty. The catchment probability maps derived from the structural realizations were compared with the well catchment forecasted by a deterministic subsurface structure, and we are able to capture this catchment within the estimated uncertainties. We believe that this study is the first to combine MPS methods with a complete data driven workflow going directly from lithological and geophysical data to realizations of the subsurface structures. The main benefits of this is that it is data driven, fast, reproducible, and transparent.

International audience; This paper compares the hydrological responses at the local scale of two models using different degrees of refinement to represent physical processes in sparsely instrumented mountainous Himalayan catchments. This work presents the novelty of applying, at a small spatio-temporal scale and under the same forcing conditions, a fully distributed surface scheme based on mass and energy balance equations (ISBA surface scheme), and a semi-distributed calibrated model (J2000 hydrological model). A new conceptual module coupled to the ISBA surface scheme for flow routing is presented. Two small catchments located in mid-and high-mountain environments were chosen to represent the very different climatic and physiographic characteristics of the Central Himalayas in the Everest region of eastern Nepal. The results show that both models globally represent the dynamic of the processes for evaporation , quick runoff and discharge in a similar way. The differences in the model structures and results mainly concern the snow processes and the soil processes. In particular for the high-mountain catchment, the snow-pack simulation is shown to be the main driver of the discrepancy between the two models. The sub-daily variations of snow processes are shown to significanlty influence the estimation of the snow-melt contribution to discharge.

Nitrogen (N) and phosphorus (P) are both essential plant nutrients. However, their joint response to plant growth is seldom described by models. This study provides an approach for modeling the joint impact of inorganic N and P fertilization on crop production, considering the P supplied by the soil, which was approximated using the soil test P (STP). We developed yield response models for Finnish spring barley crops (Hordeum vulgare L.) for clay and coarse-textured soils by using existing extensive experimental datasets and nonlinear estimation techniques. Model selection was based on iterative elimination from a wide diversity of plausible model formulations. The Cobb-Douglas type model specification, consisting of multiplicative elements, performed well against independent validation data, suggesting that the key relationships that determine crop responses are captured by the models. The estimated models were extended to dynamic economic optimization of fertilization inputs. According to the results, a fair STP level should be maintained on both coarse-textured soils (9.9 mg L-1 a(-1)) and clay soils (3.9 mg L-1 a(-1)). For coarse soils, a higher steady-state P fertilization rate is required (21.7 kg ha(-1) a(-1)) compared with clay soils (6.75 kg ha(-1) a(-1)). The steady-state N fertilization rate was slightly higher for clay soils (102.4 kg ha(-1) a(-1)) than for coarse soils (95.8 kg ha(-1) a(-1)). This study shows that the iterative elimination of plausible functional forms is a suitable method for reducing the effects of structural uncertainty on model output and optimal fertilization decisions.

Projections of the impacts of climate change on marine ecosystems are a key prerequisite for the planning of adaptation strategies, yet they are inevitably associated with uncertainty. Identifying, quantifying, and communicating this uncertainty is key to both evaluating the risk associated with a projection and building confidence in its robustness. We review how uncertainties in such projections are handled in marine science. We employ an approach developed in climate modelling by breaking uncertainty down into (i) structural (model) uncertainty, (ii) initialization and internal variability uncertainty, (iii) parametric uncertainty, and (iv) scenario uncertainty. For each uncertainty type, we then examine the current state-of-the-art in assessing and quantifying its relative importance. We consider whether the marine scientific community has addressed these types of uncertainty sufficiently and highlight the opportunities and challenges associated with doing a better job. We find that even within a relatively small field such as marine science, there are substantial differences between subdisciplines in the degree of attention given to each type of uncertainty. We find that initialization uncertainty is rarely treated explicitly and reducing this type of uncertainty may deliver gains on the seasonal-to-decadal time-scale. We conclude that all parts of marine science could benefit from a greater exchange of ideas, particularly concerning such a universal problem such as the treatment of uncertainty. Finally, marine science should strive to reach the point where scenario uncertainty is the dominant uncertainty in our projections.

Comprehensive problem framing that includes different perspectives is essential for holistic understanding of complex problems and as the first step in building models. We involved five stakeholders to frame the management problem of the Central Baltic herring fishery. By using the Bayesian belief networks (BBNs) approach, the views of the stakeholders were built into graphical influence diagrams representing variables and their dependencies. The views of the scientists involved concentrated on biological concerns, whereas the fisher, the manager, and the representative of an environmental nongovernmental organization included markets and fishing industry influences. Management measures were considered to have a relatively small impact on the development of the herring stock; their impact on socioeconomic objectives was greater. Overall, the framings by these stakeholders propose a focus on socioeconomic issues in research and management and explicitly define management objectives, not only in biological but also in social and economic terms. We find the approach an illustrative tool to structure complex issues systematically. Such a tool can be used as a forum for discussion and for decision support that explicitly includes the views of different stakeholder groups. It enables the examination of social and biological factors in one framework and facilitates bridging the gap between social and natural sciences. A benefit of the BBN approach is that the graphical model structures can be transformed into a quantitative form by inserting probabilistic information.

International audience; A general methodology is presented for the consideration of both parameter and model uncertainty in the determination of the response of geometrically nonlinear structural dynamic systems. The approach is rooted in the availability of reduced order models of these nonlinear Systems with a deterministic basis extracted from a reference model (the mean model). Uncertainty, both from parameters and model, is introduced by randomizing the coefficients of the reduced order model in a manner that guarantees the physical appropriateness of every realization of the reduced order model, i.e. while maintaining the fundamental properties of symmetry and positive definiteness of every such reduced order model. This randomization is achieved not by postulating a specific joint statistical distribution of the reduced order model coefficients but rather by deriving this distribution through the principle of maximization of the entropy constrained to satisfy the necessary symmetry and positive definiteness properties. Several desirable features of this approach are that the uncertainty can be characterized by a single measure of dispersion, affects all coefficients of the reduced order model, and is computationally easily achieved. The reduced order modeling strategy and this stochastic modeling of its coefficients are presented in details and several applications to a beam undergoing large displacement are presented. These applications demonstrate the appropriateness and computational efficiency of the method to the broad class of uncertain geometrically nonlinear dynamic systems.

International audience; This paper presents an approach for reducing uncertainty related to the process of land-cover change (LCC) prediction. LCC prediction models have, almost, two sources of uncertainty which are the uncertainty related to model parameters and the uncertainty related to model structure. These uncertainties have a big impact on decisions of the prediction model. To deal with these problems, the proposed approach is divided into three main steps: (1) an uncertainty propagation step based on possibility theory is used as a tool to evaluate the performance of the model; (2) a sensitivity analysis step based on Hartley-like measure is then used to find the most important sources of uncertainty; and (3) a knowledge base based on machine learning algorithm is built to identify the reduction factors of all uncertainty sources of parameters and to reshape their values to reduce in a significant way the uncertainty about future changes of land cover. In this study, the present and future growths of two case studies were anticipated using multi-temporal Spot-4 and Landsat satellite images. These data are used for the preparation of prediction map of year 2025. The results show that our approach based on possibility theory has a potential for reducing uncertainty in LCC prediction modeling.