Climatic drivers and biogeophysical feedbacks: a causal inference approach over multiple temporal scales

Abstract. Simulation of the carbon cycle in climate models is important due to its impact on climate change, but many weaknesses in its reproduction were found in previous models. Improvements in the representation of the land carbon cycle in Earth system models (ESMs) participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6) include the interactive treatment of both the carbon and nitrogen cycles, improved photosynthesis, and soil hydrology. To assess the impact of these model developments on aspects of the global carbon cycle, the Earth System Model Evaluation Tool (ESMValTool) is expanded to compare CO2-concentration- and CO2-emission-driven historical simulations from CMIP5 and CMIP6 to observational data sets. A particular focus is on the differences in models with and without an interactive terrestrial nitrogen cycle. Overestimations of photosynthesis (gross primary productivity (GPP)) in CMIP5 were largely resolved in CMIP6 for participating models with an interactive nitrogen cycle but remaining for models without one. This points to the importance of including nutrient limitation. Simulating the leaf area index (LAI) remains challenging, with a large model spread in both CMIP5 and CMIP6. In ESMs, the global mean land carbon uptake (net biome productivity (NBP)) is well reproduced in the CMIP5 and CMIP6 multi-model means. However, this is the result of an underestimation of NBP in the Northern Hemisphere, which is compensated by an overestimation in the Southern Hemisphere and the tropics. Carbon stocks remain a large uncertainty in the models. While vegetation carbon content is slightly better represented in CMIP6, the inter-model range of soil carbon content remains the same between CMIP5 and CMIP6. Overall, a slight improvement in the simulation of land carbon cycle parameters is found in CMIP6 compared to CMIP5, but with many biases remaining, further improvements of models in particular for LAI and NBP is required. Models from modeling groups participating in both CMIP phases generally perform similarly or better in their CMIP6 compared to their CMIP5 models. This improvement is not as significant in the multi-model means due to more new models in CMIP6, especially those using older versions of the Community Land Model (CLM). Emission-driven simulations perform just as well as the concentration-driven models, despite the added process realism. Due to this, we recommend that ESMs in future Coupled Model Intercomparison Project (CMIP) phases perform emission-driven simulations as the standard so that climate–carbon cycle feedbacks are fully active. The inclusion of the nitrogen limitation led to a large improvement in photosynthesis compared to models not including this process, suggesting the need to view the nitrogen cycle as a necessary part of all future carbon cycle models. Possible benefits when including further limiting nutrients such as phosphorus should also be considered.

An assessment is made of the modes of interannual variability in the seasonal mean summer and winter Southern Hemisphere (SH) 500-hPa geopotential height in the twentieth century in models from the Coupled Model Intercomparison Project (CMIP) phase 5 (CMIP5) dataset. Modes of variability of both the slow (signal) and intraseasonal (noise) components in the CMIP5 models are evaluated against those estimated from reanalysis data. There is general improvement in the leading modes of the slow (signal) component in CMIP5 models compared with the CMIP phase 3 (CMIP3) dataset. The largest improvement is in the spatial structures of the modes related to El Niño–Southern Oscillation variability in SH summer. An overall score metric is significantly higher for CMIP5 over CMIP3 in both seasons. The leading modes in the intraseasonal noise component are generally well reproduced in CMIP5 models, and there are few differences from CMIP3. A new total overall score metric is used to rank the CMIP5 models over both seasons. Weighting the seasons by the relative spread of overall scores is shown to be suitable for generating multimodel ensembles for further analysis of interannual variability. In multimodel ensembles, it is found that an ensemble of size 5 or 6 is sufficient in SH summer to reproduce well the dominant modes. In contrast, about 13 models are typically are required in SH winter. It is shown that it is necessary that the selected models individually reproduce well the leading modes of the slow component.

Abstract. Forests are an important component in the framework of nature-based solutions for mitigating climate change. However, there are still uncertainties about the biogeophysical effects of forest cover changes affecting heat and water fluxes as captured by Earth System Models (ESMs) simulations and observations. In this study, we investigate the differences in the surface temperature response to idealized, complete deforestation and the temperature sensitivity to percentage change in forest cover in ESMs and observations. In this comparison, the separation between local (at the place of deforestation) and non-local (nearby or distant locations) effects is crucial as observations capture only the former. Here, we propose a modified methodology to separate local and non-local effects in climate models suitable for simulations with linear rate of deforestation. The local sensitivity of a climate variable per unit deforested area is represented by the slope of the linear regression, where tree cover is an explanatory variable. The non-local effect is defined as the difference between the overall change in the respective climate variable and the local effect. Our analysis of eleven ESMs of the Coupled Model Intercomparison Project Phase 6 (CMIP6) that participated in the idealized global deforestation experiment deforest-glob, reveals a coherent local temperature response among climate models characterized by warming in the tropics and cooling in the northern higher latitudes. The temperature response however varies in magnitude, space and time with ESMs showing distinctive seasonal and spatial patterns. A closer look at the albedo response to deforestation across northern latitudes shows an overestimation in the ESMs in comparison to observations that translates via an emergent constraint (i.e. resulting from the linear relationships between local albedo and surface temperature within the model ensemble) into an overestimation of the overall simulated cooling effect. The overestimation of the local albedo sensitivity cannot be explained solely by the higher percentage of snow cover in ESMs. In terms of local latent heat flux sensitivity, the ESMs ensemble mean is overestimated for the boreal region, but it is in good agreement with the observational constraint in the temperate forests and the tropics. However, the inter-model spread and the internal model variation in these regions are considerable. ESMs having higher local albedo and latent heat flux sensitivities than the current observational constraints can still exhibit a realistic temperature response due to compensatory effects between the two sensitivities. Non-local effects contribute to consistent cooling throughout the globe, which persists also during the summer when the influence of the overestimated albedo sensitivity over snow is weaker. Having a deeper understanding of how local and non-local biogeophysical effects are represented in ESMs can give us insights into the net climate impact of deforestation and help us improve next generation ESMs.

Global evaluation of simulated surface shortwave radiation in CMIP6 models

Abstract. The response of the global climate–carbon-cycle system to anthropogenic perturbations happens differently at different timescales. The unravelling of the memory structure underlying this timescale dependence is a major challenge in climate research. Recently the widely applied α–β–γ framework proposed by Friedlingstein et al. (2003) to quantify climate–carbon-cycle feedbacks has been generalized to account also for such internal memory. By means of this generalized framework, we investigate the timescale dependence of the airborne fraction for a set of Earth system models that participated in CMIP5 (Coupled Model Intercomparison Project Phase 5). The analysis is based on published simulation data from C4MIP-type (Coupled Climate–Carbon Cycle Model Intercomparison) experiments with these models. Independently of the considered scenario, the proposed generalization describes at global scale the reaction of the climate–carbon system to sufficiently weak perturbations. One prediction from this theory is how the timescale-resolved airborne fraction depends on the underlying feedbacks between climate and the carbon cycle. These feedbacks are expressed as timescale-resolved functions depending solely on analogues of the α, β, and γ sensitivities, introduced in the generalized framework as linear response functions. In this way a feedback-dependent quantity (airborne fraction) is predicted from feedback-independent quantities (the sensitivities). This is the key relation underlying our study. As a preparatory step, we demonstrate the predictive power of the generalized framework exemplarily for simulations with the Max Planck Institute (MPI) Earth System Model. The whole approach turns out to be valid for perturbations of up to an about 100 ppm CO2 rise above the pre-industrial level; beyond this value the response becomes non-linear. By means of the generalized framework we then derive the timescale dependence of the airborne fraction from the underlying climate–carbon-cycle feedbacks for an ensemble of CMIP5 models. Our analysis reveals that for all studied CMIP5 models (1) the total climate–carbon-cycle feedback is negative at all investigated timescales, (2) the airborne fraction generally decreases for increasing timescales, and (3) the land biogeochemical feedback dominates the model spread in the airborne fraction at all these timescales. Qualitatively similar results were previously found by employing the original α–β–γ framework to particular perturbation scenarios, but our study demonstrates that, although obtained from particular scenario simulations, they are characteristics of the coupled climate–carbon-cycle system as such, valid at all considered timescales. These more general conclusions are obtained by accounting for the internal memory of the system as encoded in the generalized sensitivities, which in contrast to the original α, β, and γ are scenario-independent.

Abstract. Systematic evaluation of the carbon cycle physical and biological variables simulated in Earth System Model (ESM) participating in the Coupled Model Intercomparison Project Phase 6 (CMIP 6) is fundamental to the understanding of terrestrial ecosystems, as well as to future projections. Leaf Area Index (LAI), Gross Primary Productivity (GPP), Net Primary Productivity (NPP), Net Ecosystem Productivity (NEP) and Land Surface Temperature (LST) as key indicators of carbon cycle performance in ESM outputs, play a critical role in evaluating ecosystem functions. Assessing these metrics can provide valuable insights into the biases in model-simulated ecosystems and offer guidance for model improvement. In this study, we assessed the interannual trends performance of LAI, GPP, NPP, NEP and LST simulated by 12 CMIP6 ESMs during the historical period by using satellite LAI, NPP, NEP, LST and CSIF data as observations. The findings indicate that: (1) There are significant uncertainties in the overall trends and interannual variability in LAI, NPP, and LST captured by the CMIP6 ESM. Meanwhile, simulated GPP and NEP trends were lower than observations with discrepancies reaching 0.03 yr−1 for GPP and 2.46 gCm-2yr-1 for NEP. (2) Spatially, CMIP6 ESMs exhibited widespread underestimation of trends in LAI, GPP, NPP, and NEP across China. The MME underestimated these variables in 46.29 % (LAI), 43.47 % (GPP), 49.81 % (NPP), and 61.34 % (NEP) of the study area. Meanwhile, the simulated LST trend is underestimated in northern China, while its overestimations in western and southern China. (3) ESMs inadequate responsiveness to anthropogenic and environmental forcing and incomplete mechanistic representation of plant respiration pathways struggled accurate simulation of trends in LAI, GPP, NPP, NEP and LST.

Evaluation and future projections of wind energy resources over the Northern Hemisphere in CMIP5 and CMIP6 models

Advances in weather and climate research have demonstrated the role of the stratosphere in the Earth system across a wide range of temporal and spatial scales. Stratospheric ozone loss has been identified as a key driver of Southern Hemisphere tropospheric circulation trends, affecting ocean currents and carbon uptake, sea ice, and possibly even the Antarctic ice sheets. Stratospheric variability has also been shown to affect short-term and seasonal forecasts, connecting the tropics and midlatitudes and guiding storm-track dynamics. The two-way interactions between the stratosphere and the Earth system have motivated the World Climate Research Programme's (WCRP) Stratospheric Processes and their Role in Climate's (SPARC) activity on Modelling the Dynamics and Variability of the Stratosphere-Troposphere System (DynVar) to investigate the impact of stratospheric dynamics and variability on climate. This assessment will be made possible by two new multimodel datasets. First, roughly 10 models with a well-resolved stratosphere are participating in the Coupled Model Intercomparison Project phase 5 (CMIP5), providing the first multimodel ensemble of climate simulations coupled from the stratopause to the sea floor. Second, the Stratosphere Resolving Historical Forecast Project (Strat-HFP) of WCRP's Climate Variability and Predictability (CLIVAR) program is forming a multimodel set of seasonal hind-casts with stratosphere-resolving models, revealing the impact of both stratospheric initial conditions and dynamics on intraseasonal prediction. The CMIP5 and Strat-HFP model datasets will offer an unprecedented opportunity to understand the role of the stratosphere in the natural and forced variability of the Earth system and to determine whether incorporating knowledge of the middle atmosphere improves seasonal forecasts and climate projections.

Seventeen models participating in the Coupled Model Intercomparison Project phase 5 (CMIP5) activity are compared on their historical simulation of the South China Sea (SCS) ocean heat content (OHC) in the upper 300 m. Ishii’s temperature data, based on the World Ocean Database 2005 (WOD05) and World Ocean Atlas 2005 (WOA05), is used to assess the model performance by comparing the spatial patterns of seasonal OHC anomaly (OHCa) climatology, OHC climatology, monthly OHCa climatology, and interannual variability of OHCa. The spatial patterns in Ishii’s data set show that the seasonal SCS OHCa climatology, both in winter and summer, is strongly affected by the wind stress and the current circulations in the SCS and its neighboring areas. However, the CMIP5 models present rather different spatial patterns and only a few models properly capture the dominant features in Ishii’s pattern. Among them, GFDL-ESM2G is of the best performance. The SCS OHC climatology in the upper 300 m varies greatly in different models. Most of them are much greater than those calculated from Ishii’s data. However, the monthly OHCa climatology in each of the 17 CMIP5 models yields similar variation and magnitude as that in Ishii’s. As for the interannual variability, the standard deviations of the OHCa time series in most of the models are somewhat larger than those in Ishii’s. The correlation between the interannual time series of Ishii’s OHCa and that from each of the 17 models is not satisfactory. Among them, BCC-CSM1.1 has the highest correlation to Ishii’s, with a coefficient of about 0.6.

The rapid expansion of Earth system model (ESM) data available from the Coupled Model Intercomparison Project Phase 6 (CMIP6) necessitates new methods to evaluate the performance and suitability of ESMs used for hydroclimate applications as these extremely large data volumes complicate stakeholder efforts to use new ESM outputs in updated climate vulnerability and impact assessments. We develop an analysis framework to inform ESM sub‐selection based on process‐oriented considerations and demonstrate its performance for a regional application in the US Pacific Northwest. First, a suite of global and regional metrics is calculated, using multiple historical observation datasets to assess ESM performance. These metrics are then used to rank CMIP6 models, and a culled ensemble of models is selected using a trend‐related diagnostics approach. This culling strategy does not dramatically change climate scenario trend projections in this region, despite retaining only 20% of the CMIP6 ESMs in the final model ensemble. The reliability of the culled trend projection envelope and model response similarity is also assessed using a perfect model framework. The absolute difference in temperature trend projections is reduced relative to the full ensemble compared to the model for each SSP scenario, while precipitation trend errors are largely unaffected. In addition, we find that the spread of the culled ensemble temperature and precipitation trends includes the trend of the “truth” model ∼83%‐92% of the time. This analysis demonstrates a reliable method to reduce ESM ensemble size that can ease use of ESMs for creating and understanding climate vulnerability and impact assessments.

Abstract. The uptake of anthropogenic carbon (Cant) by the ocean leads to ocean acidification, causing the reduction of pH and the saturation states of aragonite (Ωarag) and calcite (Ωcalc). The Arctic Ocean is particularly vulnerable to ocean acidification due to its naturally low pH and saturation states and due to ongoing freshening and the concurrent reduction in total alkalinity in this region. Here, we analyse ocean acidification in the Arctic Ocean over the 21st century across 14 Earth system models (ESMs) from the latest Coupled Model Intercomparison Project Phase 6 (CMIP6). Compared to the previous model generation (CMIP5), models generally better simulate maximum sea surface densities in the Arctic Ocean and consequently the transport of Cant into the Arctic Ocean interior, with simulated historical increases in Cant in improved agreement with observational products. Moreover, in CMIP6 the inter-model uncertainty of projected changes over the 21st century in Arctic Ocean Ωarag and Ωcalc averaged over the upper 1000 m is reduced by 44–64 %. The strong reduction in projection uncertainties of Ωarag and Ωcalc can be attributed to compensation between Cant uptake and total alkalinity reduction in the latest models. Specifically, ESMs with a large increase in Arctic Ocean Cant over the 21st century tend to simulate a relatively weak concurrent freshening and alkalinity reduction, while ESMs with a small increase in Cant simulate a relatively strong freshening and concurrent total alkalinity reduction. Although both mechanisms contribute to Arctic Ocean acidification over the 21st century, the increase in Cant remains the dominant driver. Even under the low-emissions Shared Socioeconomic Pathway 1-2.6 (SSP1-2.6), basin-wide averaged Ωarag undersaturation in the upper 1000 m occurs before the end of the century. While under the high-emissions pathway SSP5-8.5, the Arctic Ocean mesopelagic is projected to even become undersaturated with respect to calcite. An emergent constraint identified in CMIP5 which relates present-day maximum sea surface densities in the Arctic Ocean to the projected end-of-century Arctic Ocean Cant inventory is found to generally hold in CMIP6. However, a coincident constraint on Arctic declines in Ωarag and Ωcalc is not apparent in the new generation of models. This is due to both the reduction in Ωarag and Ωcalc projection uncertainty and the weaker direct relationship between projected changes in Arctic Ocean Cant and changes in Ωarag and Ωcalc.

Warm sea-surface temperature (SST) biases in the southeastern tropical Atlantic (SETA), which is defined by a region from 5°E to the west coast of southern Africa and from 10°S to 30°S, are a common problem in many current and previous generation climate models. The Coupled Model Intercomparison Project Phase 5 (CMIP5) ensemble provides a useful framework to tackle the complex issues concerning causes of the SST bias. In this study, we tested a number of previously proposed mechanisms responsible for the SETA SST bias and found the following results. First, the multi-model ensemble mean shows a positive shortwave radiation bias of ~20 W m−2, consistent with models’ deficiency in simulating low-level clouds. This shortwave radiation error, however, is overwhelmed by larger errors in the simulated surface turbulent heat and longwave radiation fluxes, resulting in excessive heat loss from the ocean. The result holds for atmosphere-only model simulations from the same multi-model ensemble, where the effect of SST biases on surface heat fluxes is removed, and is not sensitive to whether the analysis region is chosen to coincide with the maximum warm SST bias along the coast or with the main SETA stratocumulus deck away from the coast. This combined with the fact that there is no statistically significant relationship between simulated SST biases and surface heat flux biases among CMIP5 models suggests that the shortwave radiation bias caused by poorly simulated low-level clouds is not the leading cause of the warm SST bias. Second, the majority of CMIP5 models underestimate upwelling strength along the Benguela coast, which is linked to the unrealistically weak alongshore wind stress simulated by the models. However, a correlation analysis between the model simulated vertical velocities and SST biases does not reveal a statistically significant relationship between the two, suggesting that the deficient coastal upwelling in the models is not simply related to the warm SST bias via vertical heat advection. Third, SETA SST biases in CMIP5 models are correlated with surface and subsurface ocean temperature biases in the equatorial region, suggesting that the equatorial temperature bias remotely contributes to the SETA SST bias. Finally, we found that all CMIP5 models simulate a southward displaced Angola–Benguela front (ABF), which in many models is more than 10° south of its observed location. Furthermore, SETA SST biases are most significantly correlated with ABF latitude, which suggests that the inability of CMIP5 models to accurately simulate the ABF is a leading cause of the SETA SST bias. This is supported by simulations with the oceanic component of one of the CMIP5 models, which is forced with observationally derived surface fluxes. The results show that even with the observationally derived surface atmospheric forcing, the ocean model generates a significant warm SST bias near the ABF, underlining the important role of ocean dynamics in SETA SST bias problem. Further model simulations were conducted to address the impact of the SETA SST biases. The results indicate a significant remote influence of the SETA SST bias on global model simulations of tropical climate, underscoring the importance and urgency to reduce the SETA SST bias in global climate models.

Ozone changes and associated climate impacts in the Coupled Model Intercomparison Project Phase 5 (CMIP5) simulations are analyzed over the historical (1960–2005) and future (2006–2100) period under four Representative Concentration Pathways (RCP). In contrast to CMIP3, where half of the models prescribed constant stratospheric ozone, CMIP5 models all consider past ozone depletion and future ozone recovery. Multimodel mean climatologies and long‐term changes in total and tropospheric column ozone calculated from CMIP5 models with either interactive or prescribed ozone are in reasonable agreement with observations. However, some large deviations from observations exist for individual models with interactive chemistry, and these models are excluded in the projections. Stratospheric ozone projections forced with a single halogen, but four greenhouse gas (GHG) scenarios show largest differences in the northern midlatitudes and in the Arctic in spring (~20 and 40 Dobson units (DU) by 2100, respectively). By 2050, these differences are much smaller and negligible over Antarctica in austral spring. Differences in future tropospheric column ozone are mainly caused by differences in methane concentrations and stratospheric input, leading to ~10 DU increases compared to 2000 in RCP 8.5. Large variations in stratospheric ozone particularly in CMIP5 models with interactive chemistry drive correspondingly large variations in lower stratospheric temperature trends. The results also illustrate that future Southern Hemisphere summertime circulation changes are controlled by both the ozone recovery rate and the rate of GHG increases, emphasizing the importance of simulating and taking into account ozone forcings when examining future climate projections.

Assessing the regional and local impacts of climate changes on rainfall and streamflow patterns is often challenging, as most results tend to be very dispersed and inconsistent across models. We offer a complementary alternative for climate change impact studies by focusing primarily on climate indices. In particular, we analyse the historical and future variability of key climate indices associated with the South America Monsoon System (SAMS) as simulated by Coupled Model Intercomparison Project—Phase 5 (CMIP5) models. We use climate indices related to the tropical Atlantic interhemispheric surface temperature gradient, associated with the meridional position of the Intertropical Convergence Zone (ITCZ), as well as the land‐ocean temperature contrast in the tropical Atlantic‐South America region, and the intensity and position of the South Atlantic Subtropical High (SASH). We selected 14 CMIP5 models to estimate climate indices for the retrospective period (1980–2005) and future period (2006–2099), the latter under the RCP 4.5 and RCP 8.5 scenarios. For comparison purposes during the retrospective period, the indices were also calculated using data from the Era‐Interim reanalysis. The seasonal behaviour of the Era‐Interim based indices is well reproduced by CMIP5 models, although the magnitude is significantly higher. For the future period, most models under both scenarios showed a consistent and asymmetrical intensification of the dipole index, suggesting, in a future climate, a potential wider displacement of the ITCZ during its seasonal cycle. As for the land‐ocean contrast index, the CMIP5 simulations indicated consistent increase in the magnitude for the positive phase (land warmer than ocean) and a decrease for the negative phase (land colder than ocean). Finally, CMIP5 models consistently predict a strengthening and southward displacement of the SASH centre. To the best of our knowledge, this is the first study to reveal future changes in climate indices associated with the SAMS that shows relevant coherences across models and scenarios.

Previous studies have hypothesized that climatologically thick salinity-stratified Barrier Layers (BL) in the North Indian Ocean (NIO) could influence the upper ocean heat budget, sea surface temperature (SST) and monsoon. Here, we investigate the performance of state-of-the-art climate models from the Coupled Model Intercomparison Project phase 6 (CMIP6) in simulating the barrier layer thickness (BLT) in the NIO. CMIP6 models generally reproduce the main features of the BLT seasonal cycle and spatial distribution, but with a shallow November-February (NDJF) BLT bias in regions with thick observed BLT (eastern equatorial Indian Ocean [EEIO], Bay of Bengal [BoB] and southeastern Arabian Sea [SEAS]). CMIP6 models display an easterly equatorial zonal surface wind bias linked to dry rainfall and cold SST biases in the southern BoB, through the Bjerknes feedback loop. The easterly equatorial bias is also responsible for the shallow isothermal layer depth (ILD) and BLT bias in the EEIO. The underestimated rainfall over the BoB leads to higher sea surface salinity (SSS) and too deep mixed layer depth (MLD), resulting in the BoB BLT bias. The intensity of the easterly equatorial bias also contributes to the inter-model spread in BoB BLT bias, through the propagation of EEIO ILD signals into the coastal waveguide. Finally, the SEAS BLT bias is due to a too deep MLD, which is predominantly controlled by the high SSS related to attenuated monsoonal currents around India and a reduced inflow of BoB low-salinity water. The BL effect on the mixed layer entrainment cooling does not seem to operate in CMIP6 simulations. Rather, deep salinity-related MLD biases in the BoB result in a diminished cooling rate in response to winter negative surface heat fluxes, and hence alleviate cold BoB SST biases. This suggests that salinity effects alleviate the biases that develop through the positive Bejrknes feedback loop between BoB SST, BoB rainfall and equatorial wind stresses in CMIP6.