Fitting individual-based models of spatial population dynamics to long-term monitoring data.

Generating spatial predictions of species distribution is a central task for research and policy. Currently, correlative species distribution models (cSDMs) are among the most widely used tools for this purpose. However, a fundamental assumption of cSDMs, that species distributions are in equilibrium with their environment, is rarely fulfilled in real data and limits the applicability of cSDMs for dynamic projections. Process-based, dynamic SDMs (dSDMs) promise to overcome these limitations as they explicitly represent transient dynamics and enhance spatiotemporal transferability. Software tools for implementing dSDMs are becoming increasingly available, but their parameter estimation can be complex. Here, we test the feasibility of calibrating and validating a dSDM using long-term monitoring data of Swiss red kites (Milvus milvus). This population has shown strong increases in abundance and a progressive range expansion over the last decades, indicating a nonequilibrium situation. We construct an individual-based model using the RangeShiftR modeling platform and use Bayesian inference for model calibration. This allows the integration of heterogeneous data sources, such as parameter estimates from published literature and observational data from monitoring schemes, with a coherent assessment of parameter uncertainty. Our monitoring data encompass counts of breeding pairs at 267 sites across Switzerland over 22 years. We validate our model using a spatial-block cross-validation scheme and assess predictive performance with a rank-correlation coefficient. Our model showed very good predictive accuracy of spatial projections and represented well the observed population dynamics over the last two decades. Results suggest that reproductive success was a key factor driving the observed range expansion. According to our model, the Swiss red kite population fills large parts of its current range but has potential for further increases in density. We demonstrate the practicality of data integration and validation for dSDMs using RangeShiftR. This approach can improve predictive performance compared to cSDMs. The workflow presented here can be adopted for any population for which some prior knowledge on demographic and dispersal parameters as well as spatiotemporal observations of abundance or presence/absence are available. The fitted model provides improved quantitative insights into the ecology of a species, which can greatly aid conservation and management efforts.

Novel community data in ecology-properties and prospects.

New technologies for monitoring biodiversity such as environmental (e)DNA, passive acoustic monitoring, and optical sensors promise to generate automated spatiotemporal community observations at unprecedented scales and resolutions. Here, we introduce 'novel community data' as an umbrella term for these data. We review the emerging field around novel community data, focusing on new ecological questions that could be addressed; the analytical tools available or needed to make best use of these data; and the potential implications of these developments for policy and conservation. We conclude that novel community data offer many opportunities to advance our understanding of fundamental ecological processes, including community assembly, biotic interactions, micro- and macroevolution, and overall ecosystem functioning.

Accuracy, realism and general applicability of European forest models.

Forest models are instrumental for understanding and projecting the impact of climate change on forests. A considerable number of forest models have been developed in the last decades. However, few systematic and comprehensive model comparisons have been performed in Europe that combine an evaluation of modelled carbon and water fluxes and forest structure. We evaluate 13 widely used, state-of-the-art, stand-scale forest models against field measurements of forest structure and eddy-covariance data of carbon and water fluxes over multiple decades across an environmental gradient at nine typical European forest stands. We test the models' performance in three dimensions: accuracy of local predictions (agreement of modelled and observed annual data), realism of environmental responses (agreement of modelled and observed responses of daily gross primary productivity to temperature, radiation and vapour pressure deficit) and general applicability (proportion of European tree species covered). We find that multiple models are available that excel according to our three dimensions of model performance. For the accuracy of local predictions, variables related to forest structure have lower random and systematic errors than annual carbon and water flux variables. Moreover, the multi-model ensemble mean provided overall more realistic daily productivity responses to environmental drivers across all sites than any single individual model. The general applicability of the models is high, as almost all models are currently able to cover Europe's common tree species. We show that forest models complement each other in their response to environmental drivers and that there are several cases in which individual models outperform the model ensemble. Our framework provides a first step to capturing essential differences between forest models that go beyond the most commonly used accuracy of predictions. Overall, this study provides a point of reference for future model work aimed at predicting climate impacts and supporting climate mitigation and adaptation measures in forests.

Studying speciation and extinction dynamics from phylogenies: addressing identifiability issues.

A lot of what we know about past speciation and extinction dynamics is based on statistically fitting birth-death processes to phylogenies of extant species. Despite their wide use, the reliability of these tools is regularly questioned. It was recently demonstrated that vast 'congruent' sets of alternative diversification histories cannot be distinguished (i.e., are not identifiable) using extant phylogenies alone, reanimating the debate about the limits of phylogenetic diversification analysis. Here, we summarize what we know about the identifiability of the birth-death process and how identifiability issues can be addressed. We conclude that extant phylogenies, when combined with appropriate prior hypotheses and regularization techniques, can still tell us a lot about past diversification dynamics.

Climate-driven, but dynamic and complex? A reconciliation of competing hypotheses for species' distributions.

Estimates of the percentage of species "committed to extinction" by climate change range from 15% to 37%. The question is whether factors other than climate need to be included in models predicting species' range change. We created demographic range models that include climate vs. climate-plus-competition, evaluating their influence on the geographic distribution of Pinus edulis, a pine endemic to the semiarid southwestern U.S. Analyses of data on 23,426 trees in 1941 forest inventory plots support the inclusion of competition in range models. However, climate and competition together only partially explain this species' distribution. Instead, the evidence suggests that climate affects other range-limiting processes, including landscape-scale, spatial processes such as disturbances and antagonistic biotic interactions. Complex effects of climate on species distributions-through indirect effects, interactions, and feedbacks-are likely to cause sudden changes in abundance and distribution that are not predictable from a climate-only perspective.

gen3sis: A general engine for eco-evolutionary simulations of the processes that shape Earth's biodiversity.

Understanding the origins of biodiversity has been an aspiration since the days of early naturalists. The immense complexity of ecological, evolutionary, and spatial processes, however, has made this goal elusive to this day. Computer models serve progress in many scientific fields, but in the fields of macroecology and macroevolution, eco-evolutionary models are comparatively less developed. We present a general, spatially explicit, eco-evolutionary engine with a modular implementation that enables the modeling of multiple macroecological and macroevolutionary processes and feedbacks across representative spatiotemporally dynamic landscapes. Modeled processes can include species' abiotic tolerances, biotic interactions, dispersal, speciation, and evolution of ecological traits. Commonly observed biodiversity patterns, such as α, ß, and γ diversity, species ranges, ecological traits, and phylogenies, emerge as simulations proceed. As an illustration, we examine alternative hypotheses expected to have shaped the latitudinal diversity gradient (LDG) during the Earth's Cenozoic era. Our exploratory simulations simultaneously produce multiple realistic biodiversity patterns, such as the LDG, current species richness, and range size frequencies, as well as phylogenetic metrics. The model engine is open source and available as an R package, enabling future exploration of various landscapes and biological processes, while outputs can be linked with a variety of empirical biodiversity patterns. This work represents a key toward a numeric, interdisciplinary, and mechanistic understanding of the physical and biological processes that shape Earth's biodiversity.

Siberian plants shift their phenology in response to climate change.

Siberia has undergone dramatic climatic changes due to global warming in recent decades. Yet, the ecological responses to these climatic changes are still poorly understood due to a lack of data. Here, we use a unique data set from the Russian 'Chronicles of Nature' network to analyse the long-term (1976-2018) phenological shifts in leaf out, flowering, fruiting and senescence of 67 common Siberian plant species. We find that Siberian boreal forest plants advanced their early season (leaf out and flowering) and mid-season (fruiting) phenology by -2.2, -0.7 and -1.6 days/decade, and delayed the onset of senescence by 1.6 days/decade during this period. These mean values, however, are subject to substantial intraspecific variability, which is partly explained by the plants' growth forms. Trees and shrubs advanced leaf out and flowering (-3.1 and -3.3. days/decade) faster than herbs (-1 day/decade), presumably due to the more direct exposure of leaf and flower buds to ambient air for the woody vegetation. For senescence, we detected a reverse pattern: stronger delays in herbs (2.1 days/decade) than in woody plants (1.0-1.2 days/decade), presumably due to the stronger effects of autumn frosts on the leaves of herbs. Interestingly, the timing of fruiting in all four growth forms advanced at similar paces, from 1.4 days/decade in shrubs to 1.7 days/decade in trees and herbs. Our findings point to a strong, yet heterogeneous, response of Siberian plant phenology to recent global warming. Furthermore, the results highlight that species- and growth form-specific differences among study species could be used to identify plants particularly at risk of decline due to their low adaptive capacity or a loss of synchronization with important interaction partners.

Linking functional traits and demography to model species-rich communities.

It has long been anticipated that relating functional traits to species demography would be a cornerstone for achieving large-scale predictability of ecological systems. If such a relationship existed, species demography could be modeled only by measuring functional traits, transforming our ability to predict states and dynamics of species-rich communities with process-based community models. Here, we introduce a new method that links empirical functional traits with the demographic parameters of a process-based model by calibrating a transfer function through inverse modeling. As a case study, we parameterize a modified Lotka-Volterra model of a high-diversity mountain grassland with static plant community and functional trait data only. The calibrated trait-demography relationships are amenable to ecological interpretation, and lead to species abundances that fit well to the observed community structure. We conclude that our new method offers a general solution to bridge the divide between trait data and process-based models in species-rich ecosystems.

Closure or medical therapy of patent foramen ovale in cryptogenic stroke: prospective case series.

BACKGROUND: Results of randomized controlled trials (RCT) do not provide definite guidance for secondary prevention after ischemic stroke (IS)/transient ischemic attack (TIA) attributed to patent foramen ovale (PFO). No recommendations can be made for patients > 60 years. We aimed to compare interventional and medical PFO-management in cryptogenic IS/TIA patients, including patients > 60 years. METHODS: Prospective case series including consecutive cryptogenic IS/TIA patients with PFO at Tuebingen university stroke unit, Germany. 'PFO-closure' was recommended in patients ≤70 years when featuring high-risk PFO (i.e., with atrial septal aneurysm, spontaneous, or high-grade right-to-left shunt during Valsalva). Primary (recurrent IS/intracranial hemorrhage) and secondary endpoints (e.g., disability) were assessed during ≥1-year follow-up; planned subgroup analyses of patients ≤60/> 60 years. RESULTS: Among 236 patients with median age of 58 (range 18-88) years, 38.6% were females and median presenting National Institutes of Health Stroke Scale score was 1 (IQR 0-4). Mean follow-up was 2.8 ± 1.3 years. No intracranial hemorrhage was observed. Recurrent IS rate after 'PFO-closure' was 2.9% (95%CI 0-6.8%) and 7% (4-16.4) in high-risk PFO patients ≤60 (n = 103) and > 60 years (n = 43), respectively, versus 4% (0-11.5) during 'medical therapy alone' MTA (n = 28). 42 low-risk PFO patients treated with MTA experienced no recurrent IS/TIA. CONCLUSIONS: In our real-world study, IS recurrence rate in 'PFO-closure' high-risk PFO patients ≤60 years was comparable to that observed in recent RCT. High-risk PFO patients > 60 years who underwent PFO-closure had similar IS recurrence rates than those who received MTA. MTA seems the appropriate treatment for low-risk PFO. TRIAL REGISTRATION: ClinicalTrials.gov, registration number: NCT04352790 , registered on: April 20, 2020 - retrospectively registered.

Towards robust statistical inference for complex computer models.

Ecologists increasingly rely on complex computer simulations to forecast ecological systems. To make such forecasts precise, uncertainties in model parameters and structure must be reduced and correctly propagated to model outputs. Naively using standard statistical techniques for this task, however, can lead to bias and underestimation of uncertainties in parameters and predictions. Here, we explain why these problems occur and propose a framework for robust inference with complex computer simulations. After having identified that model error is more consequential in complex computer simulations, due to their more pronounced nonlinearity and interconnectedness, we discuss as possible solutions data rebalancing and adding bias corrections on model outputs or processes during or after the calibration procedure. We illustrate the methods in a case study, using a dynamic vegetation model. We conclude that developing better methods for robust inference of complex computer simulations is vital for generating reliable predictions of ecosystem responses.

Point-of-care testing for emergency assessment of coagulation in patients treated with direct oral anticoagulants including edoxaban.

BACKGROUND: Direct oral anticoagulants (DOAC) including edoxaban are increasingly used for stroke prevention in atrial fibrillation. Despite treatment, annual stroke rate in these patients remains 1-2%. Rapid assessment of coagulation would be useful to guide thrombolysis or reversal therapy in this growing population of DOAC/edoxaban-treated stroke patients. Employing the Hemochron™ Signature Elite point-of-care test system (HC-POCT), clinically relevant plasma concentrations of dabigatran and rivaroxaban can be excluded in a blood sample. However, no data exists on the effect of edoxaban on HC-POCT results. We evaluated whether edoxaban plasma concentrations above the current treatment thresholds for thrombolysis or anticoagulation reversal (i.e., 30 and 50 ng/mL) can be ruled out with the HC-POCT. METHODS: We prospectively studied patients receiving a first dose of edoxaban. Six blood samples were collected from each patient: before, 0.5, 1, 2, 8, and 24 h after drug intake. HC-POCT-based INR (HC-INR), activated clotting time (HC-ACT+ and HC-ACT-LR), activated partial thromboplastin time (HC-aPTT), and mass spectrometry for edoxaban plasma concentrations were performed at each time-point. We calculated correlations, receiver operating characteristics (ROC) and test-specific cut-offs for ruling out edoxaban concentrations > 30 and > 50 ng/mL in a blood sample. RESULTS: One hundred twenty blood samples from 20 edoxaban-treated patients were analyzed. Edoxaban plasma concentrations ranged from 0 to 512 ng/mL. HC-INR/HC-ACT+/HC-ACT-LR/HC-aPTT ranged from 0.7-8.3/78-310 s/65-215 s/19-93 s, and Pearson's correlation coefficients showed moderate to very strong correlations with edoxaban concentrations (r = 0.95/0.79/0.70/0.60). With areas under the ROC curve of 0.997 (95% confidence interval: 0.991-0.971) and 0.989 (0.975-1.000), HC-INR most reliably ruled out edoxaban concentrations > 30 and > 50 ng/mL, respectively, and HC-INR results ≤1.5 and ≤ 2.1 provided specificity/sensitivity of 98.6% (91.2-99.9)/98.0% (88.0-99.9) and 96.8% (88.0-99.4)/96.5% (86.8-99.4). CONCLUSIONS: Our study represents the first systematic evaluation of the HC-POCT in edoxaban-treated patients. Applying sufficiently low assay-specific cut-offs, the HC-POCT may not only be used to reliably rule out dabigatran and rivaroxaban, but also very low edoxaban concentrations in a blood sample. Because the assay-specific cut-offs were retrospectively defined, further investigation is warranted. TRIAL REGISTRATION: ClinicalTrials.gov, registration number: NCT02825394 , registered on: 07/07/2016, URL.

Is Variation in Conspecific Negative Density Dependence Driving Tree Diversity Patterns at Large Scales?

Half a century ago, Janzen and Connell hypothesized that the high tree species diversity in tropical forests is maintained by specialized natural enemies. Along with other mechanisms, these can cause conspecific negative density dependence (CNDD) and thus maintain species diversity. Numerous studies have measured proxies of CNDD worldwide, but doubt about its relative importance remains. We find ample evidence for CNDD in local populations, but methodological limitations make it difficult to assess if CNDD scales up to control community diversity and thereby local and global biodiversity patterns. A combination of more robust statistical methods, new study designs, and eco-evolutionary models are needed to provide a more definite evaluation of the importance of CNDD for geographic variation in plant species diversity.

Specific Point-of-Care Testing of Coagulation in Patients Treated with Dabigatran.

BACKGROUND AND PURPOSE: Accurate and rapid assessment of coagulation status is necessary to guide thrombolysis or reversal of anticoagulation in stroke patients, but commercially available point-of-care (POC) assays are not suited for coagulation testing in patients treated with direct oral anticoagulants (DOACs). We aimed to evaluate the direct thrombin monitoring (DTM) test card by Helena Laboratories (Texas, United States) for anti-IIa-specific POC coagulation testing, hypothesizing that its POC-ecarin clotting time (POC-ECT) accurately reflects dabigatran plasma concentrations. METHODS: A prospective single-center diagnostic study (ClinicalTrials.gov-identifier: NCT02825394) was conducted enrolling patients receiving a first dose of dabigatran and patients already on dabigatran treatment. Blood samples were collected before drug intake and 0.5, 1, 2, 8, and 12 hours after intake. POC-ECT was performed using whole blood (WB), citrated blood (CB), and citrated plasma (CP). Dabigatran plasma concentrations were determined by mass spectrometry. RESULTS: In total, 240 blood samples from 40 patients contained 0 to 275 ng/mL of dabigatran. POC-ECT with WB/CB/CP ranged from 20 to 186/184/316 seconds. Pearson's correlation coefficient showed a strong correlation between dabigatran concentrations and POC-ECT with WB/CB/CP (R2 = 0.78/0.90/0.92). Dabigatran concentrations >30 and >50 ng/mL (thresholds for thrombolysis, surgery, and reversal therapy according to clinical guidelines) were detected by POC-ECT with WB/CB/CP (>36/35/45 and >43/45/59 seconds) with 95/97/97 and 96/98/97% sensitivity, and 81/87/94 and 74/60/91% specificity. CONCLUSION: This first study evaluating DOAC-specific POC coagulation testing revealed an excellent correlation of POC-ECT with actual dabigatran concentrations. Detecting clinically relevant dabigatran levels with high sensitivity/specificity, the DTM assay represents a suitable diagnostic tool in acute stroke, hemorrhage, and urgent surgery.

The Minimum Detectable Difference (MDD) Concept for Establishing Trust in Nonsignificant Results: A Critical Review.

Current regulatory guidelines for pesticide risk assessment recommend that nonsignificant results should be complemented by the minimum detectable difference (MDD), a statistical indicator that is used to decide whether the experiment could have detected biologically relevant effects. We review the statistical theory of the MDD and perform simulations to understand its properties and error rates. Most importantly, we compare the skill of the MDD in distinguishing between true and false negatives (i.e., type II errors) with 2 alternatives: the minimum detectable effect (MDE), an indicator based on a post hoc power analysis common in medical studies; and confidence intervals (CIs). Our results demonstrate that MDD and MDE only differ in that the power of the MDD depends on the sample size. Moreover, although both MDD and MDE have some skill in distinguishing between false negatives and true absence of an effect, they do not perform as well as using CI upper bounds to establish trust in a nonsignificant result. The reason is that, unlike the CI, neither MDD nor MDE consider the estimated effect size in their calculation. We also show that MDD and MDE are no better than CIs in identifying larger effects among the false negatives. We conclude that, although MDDs are useful, CIs are preferable for deciding whether to treat a nonsignificant test result as a true negative, or for determining an upper bound for an unknown true effect. Environ Toxicol Chem 2020;39:2109-2123. © 2020 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Plant species richness increases with light availability, but not variability, in temperate forests understorey.

BACKGROUND: Temperate forest understorey vegetation poses an excellent study system to investigate whether increases in resource availability lead to an increase in plant species richness. Most sunlight is absorbed by the species-poor tree canopy, making the much more species-rich understorey species inhabit a severely resource-limited habitat. Additionally, the heterogeneity of light availability, resulting from management-moderated tree composition and age structure, may contribute to species coexistence. One would therefore expect that the diversity in the herb layer correlates positively with either the overall light availability, or the light heterogeneity, depending on whether resource availability or heterogeneity are more important drivers of diversity. To test this idea, we assessed variability of light conditions in 75 forest plots across three ecoregions with four different methods. RESULTS: We correlated these data with vegetation relevés and found light availability to be strongly positively correlated with understorey plant species richness, as well as with understorey cover. Light variability (assessed with two approaches) within plots was positively correlated with transmittance, but did not improve the relationship further, suggesting that the main driver of species richness in this system is the overall resource availability. Two of the three beech-dominated regions exhibited near-identical effects of light transmittance, while the third, featuring pine alongside beech and thus with the longest gradient of transmittance and lowest species richness, displayed a weaker light response. CONCLUSIONS: While site conditions are certainly responsible for the trees selected by foresters, for the resulting forest structure, and for the differences in plant species pools, our results suggest that light transmittance is a strong mediating factor of understorey plant species richness.

Point-of-care testing of coagulation in patients treated with edoxaban.

Edoxaban, alongside other direct oral anticoagulants (DOAC), is increasingly used for prevention of thromboembolism, including stroke. Despite DOAC therapy, however, annual stroke rate in patients with atrial fibrillation remains 1-2%. Rapid exclusion of relevant anticoagulation is necessary to guide thrombolysis or reversal therapy but, so far, no data exists on the effect of edoxaban on available point-of-care test systems (POCT). To complete our previous investigation on global coagulation-POCT for the detection of DOAC, we evaluated whether CoaguChek®-INR (CC-INR) is capable of safely ruling out edoxaban concentrations above the current treatment thresholds of 30/50 ng/mL in a blood sample. We studied patients receiving a first dose of edoxaban; excluding subjects receiving other anticoagulants. Six blood samples were collected from each patient: before drug intake, 0.5, 1, 2 and 8 h after intake, and at trough (24 h). CC-INR and mass spectrometry for edoxaban concentrations were performed for each time-point. One hundred and twenty blood samples from 20 patients contained 0-302 ng/mL of edoxaban. CC-INR ranged from 0.9 to 2.3. Pearson's correlation coefficient showed strong correlation between CC-INR and edoxaban concentrations (r = 0.73, p < 0.001). Edoxaban concentrations > 30 and > 50 ng/mL were ruled out by CC-INR ≤ 1.0 and ≤ 1.1, respectively, with high specificity (> 95%), and a sensitivity of 44% (95%-confidence interval: 30-59%) and 86% (74-93%), respectively. Our study represents the first evaluation of coagulation-POCT in edoxaban-treated patients. CC-POCT is suitable to safely exclude clinically relevant edoxaban concentrations prior to thrombolysis, or guide reversal therapy in stroke patients.

Advancing an interdisciplinary framework to study seed dispersal ecology.

Although dispersal is generally viewed as a crucial determinant for the fitness of any organism, our understanding of its role in the persistence and spread of plant populations remains incomplete. Generalizing and predicting dispersal processes are challenging due to context dependence of seed dispersal, environmental heterogeneity and interdependent processes occurring over multiple spatial and temporal scales. Current population models often use simple phenomenological descriptions of dispersal processes, limiting their ability to examine the role of population persistence and spread, especially under global change. To move seed dispersal ecology forward, we need to evaluate the impact of any single seed dispersal event within the full spatial and temporal context of a plant's life history and environmental variability that ultimately influences a population's ability to persist and spread. In this perspective, we provide guidance on integrating empirical and theoretical approaches that account for the context dependency of seed dispersal to improve our ability to generalize and predict the consequences of dispersal, and its anthropogenic alteration, across systems. We synthesize suitable theoretical frameworks for this work and discuss concepts, approaches and available data from diverse subdisciplines to help operationalize concepts, highlight recent breakthroughs across research areas and discuss ongoing challenges and open questions. We address knowledge gaps in the movement ecology of seeds and the integration of dispersal and demography that could benefit from such a synthesis. With an interdisciplinary perspective, we will be able to better understand how global change will impact seed dispersal processes, and potential cascading effects on plant population persistence, spread and biodiversity.

Retinal oedema in central retinal artery occlusion develops as a function of time.

PURPOSE: Time is the key criterion in the management of non-arteritic central retinal artery occlusion (NA-CRAO). However, the precise onset of vision loss is often difficult to determine. This study aimed to evaluate the temporal changes of retinal thickness in acute NA-CRAO and the potential of this parameter to be used as a surrogate marker to estimate the onset of retinal ischaemia. METHODS: Optical coherence tomography was used to continuously assess retinal thickness and oedema progression rate in six porcine eyes. Additionally, a retrospective analysis of 12 patients with acute NA-CRAO was performed to determine association strength and progression rate between retinal thickness and onset of ischaemia. All Optical coherence tomography (OCT) scans (pigs and NA-CRAO patients) were performed within an ischaemic time frame of up to 9 hr. RESULTS: Retinal oedema progression rate in pigs was 25.32 µm/hr [CI 95%: 24.24-26.40 µm/hr]. Retrospective analysis of the patients revealed a strong correlation between retinal oedema and duration of ischaemia (Spearman's rho = 0.77, p = 0.004) with an estimated progression rate of 10.02 µm/hr [CI 95%: 3.30-16.74 µm/hr]. CONCLUSION: Retinal thickness increases with oedema formation, and ischaemia onset is strongly correlated with this structural biomarker in both, pigs and NA-CRAO patients. Prospective clinical trials will have to determine the clinical feasibility of retinal thickness measurements as a biomarker to support clinical management of NA-CRAO.

Assessing the response of forest productivity to climate extremes in Switzerland using model-data fusion.

The response of forest productivity to climate extremes strongly depends on ambient environmental and site conditions. To better understand these relationships at a regional scale, we used nearly 800 observation years from 271 permanent long-term forest monitoring plots across Switzerland, obtained between 1980 and 2017. We assimilated these data into the 3-PG forest ecosystem model using Bayesian inference, reducing the bias of model predictions from 14% to 5% for forest stem carbon stocks and from 45% to 9% for stem carbon stock changes. We then estimated the productivity of forests dominated by Picea abies and Fagus sylvatica for the period of 1960-2018, and tested for productivity shifts in response to climate along elevational gradient and in extreme years. Simulated net primary productivity (NPP) decreased with elevation (2.86 ± 0.006 Mg C ha-1  year-1  km-1 for P. abies and 0.93 ± 0.010 Mg C ha-1  year-1  km-1 for F. sylvatica). During warm-dry extremes, simulated NPP for both species increased at higher and decreased at lower elevations, with reductions in NPP of more than 25% for up to 21% of the potential species distribution range in Switzerland. Reduced plant water availability had a stronger effect on NPP than temperature during warm-dry extremes. Importantly, cold-dry extremes had negative impacts on regional forest NPP comparable to warm-dry extremes. Overall, our calibrated model suggests that the response of forest productivity to climate extremes is more complex than simple shift toward higher elevation. Such robust estimates of NPP are key for increasing our understanding of forests ecosystems carbon dynamics under climate extremes.