Generation Of Earth System Models Research Articles

Global Optimization of Soil Texture Maps From Satellite‐Observed Soil Moisture Drydowns and Its Implementation in Noah‐MP Land Surface Model

AbstractSoil moisture (SM) plays an important role in regulating regional weather and climate. However, the simulations of SM in current land surface models (LSMs) contain large biases and model spreads. One primary reason contributing to such model biases could be the misrepresentation of soil texture in LSMs, since current available large‐scale soil texture data are often generated from extrapolation algorithm based on a scarce number of in‐situ geological measurements. Fortunately, recent advancements in satellite technology provide a unique opportunity to constrain the soil texture data sets by introducing observed information at large spatial scales. Here, two major soil texture baseline data sets (Global Soil Data sets for Earth system science, GSDE and Harmonized World Soil Data from Food and Agriculture Organization, HWSD) are optimized with satellite‐estimated soil hydraulic parameters. The optimized soil maps show increased (decreased) sand (clay) content over arid regions. The soil organic carbon (SOC) content increases globally especially over regions with dense vegetation cover. The optimized soil texture data sets are then used to run simulations in one example LSM, that is, Noah LSM with Multiple Parameters. Results show that the simulated SM with satellite‐optimized soil texture maps is improved at both grid and in‐situ scales. Intercase comparison analyses show the SM improvement differs between simulations using different soil maps and soil hydraulic schemes. Our results highlight the importance of incorporating observation‐oriented calibration on soil texture in current LSMs. This study also joins the call for a better soil profile representation in the next generation of Earth System Models (ESMs).

Spatial biases reduce the ability of Earth system models to simulate soil heterotrophic respiration fluxes

Abstract. Heterotrophic respiration (Rh) is, at a global scale, one of the largest CO2 fluxes between the Earth's surface and atmosphere and may increase in the future. The previous generation of Earth system models (ESMs) was able to reproduce global fluxes relatively well, but at that, time no gridded products were available to perform an in-depth evaluation. The capacity of the new generation of ESMs used within the Coupled Model Intercomparison Project Phase 6 (CMIP6) to reproduce this flux has not been evaluated, meaning that the realism of resulting CO2 flux estimates is unclear. In this study, we combine recently released observational data on Rh and ESM simulations to evaluate the ability of 13 ESMs from CMIP6 to reproduce Rh. Only 4 of the 13 tested ESMs were able to reproduce the total Rh flux, but spatial analysis underlined important bias compensation for most of the ESMs, which generally showed an overestimation in tropical regions and an underestimation in arid regions. To identify the main drivers of the bias, we performed an analysis of the residuals and found that mean annual precipitation was the most important driver explaining the difference between ESM simulations and observation-derived products of Rh, with a higher bias between ESM simulations and Rh products where precipitation was high. Based on our results, next-generation ESMs should focus on improving the response of Rh to soil moisture.

Modeling Past Hothouse Climates as a Means for Assessing Earth System Models and Improving the Understanding of Warm Climates

Simulating the warmth and equability of past hothouse climates has been a challenge since the inception of paleoclimate modeling. The newest generation of Earth system models (ESMs) has shown substantial improvements in the ability to simulate the early Eocene global mean surface temperature (GMST) and equator-to-pole gradient. Results using the Community Earth System Model suggest that parameterizations of atmospheric radiation, convection, and clouds largely determine the Eocene GMST and are responsible for improvements in the new ESMs, but they have less direct influence on the equator-to-pole temperature gradient. ESMs still have difficulty simulating some regional and seasonal temperatures, although improved data reconstructions of chronology, spatial coverage, and seasonal resolution are needed for more robust model assessment. Looking forward, key processes including radiation and clouds need to be benchmarked and improved using more accurate models of limited domain/physics. Earth system processes need to be better explored, leveraging the increasing ESM resolution and complexity. ▪Earth system models (ESMs) are now able to simulate the large-scale features of the early Eocene.▪Remaining model-data discrepancies exist at regional and seasonal scales and require improvements in both proxy data and ESMs.▪A hierarchical modeling approach is needed to ensure relevant physical processes are parameterized reasonably well in ESMs.▪Future work is needed to leverage the continuously increasing resolution and complexity of ESMs.

Zooplankton grazing is the largest source of uncertainty for marine carbon cycling in CMIP6 models

The current generation of Earth system models used by the United Nations to project future climate scenarios (CMIP6) relies heavily on marine biogeochemical models to track the fate of carbon absorbed into the oceans. Here we compare 11 CMIP6 marine biogeochemical models and find the largest source of inter-model uncertainty in their representation of the marine carbon cycle is phytoplankton-specific loss rates to zooplankton grazing. This uncertainty is over three times larger than that of net primary production and driven by large differences in prescribed zooplankton grazing dynamics. We run a controlled sensitivity experiment in a global marine biogeochemical model and find that small changes in prescribed grazing dynamics (roughly 5% of what is used across CMIP6 models) can increase secondary and export production by 5 and 2 PgC yr−1, respectively, even when tuned to identical net primary production, likely biasing predictions of future climate states and food security.

Representation of soil hydrology in permafrost regions may explain large part of inter-model spread in simulated Arctic and subarctic climate

Abstract. The current generation of Earth system models exhibits large inter-model differences in the simulated climate of the Arctic and subarctic zone, with differences in model structure and parametrizations being one of the main sources of uncertainty. One particularly challenging aspect in modelling is the representation of terrestrial processes in permafrost-affected regions, which are often governed by spatial heterogeneity far below the resolution of the models' land surface components. Here, we use the Max Planck Institute (MPI) Earth System Model to investigate how different plausible assumptions for the representation of permafrost hydrology modulate land–atmosphere interactions and how the resulting feedbacks affect not only the regional and global climate, but also our ability to predict whether the high latitudes will become wetter or drier in a warmer future. Focusing on two idealized setups that induce comparatively “wet” or “dry” conditions in regions that are presently affected by permafrost, we find that the parameter settings determine the direction of the 21st-century trend in the simulated soil water content and result in substantial differences in the land–atmosphere exchange of energy and moisture. The latter leads to differences in the simulated cloud cover during spring and summer and thus in the planetary energy uptake. The respective effects are so pronounced that uncertainties in the representation of the Arctic hydrological cycle can help to explain a large fraction of the inter-model spread in regional surface temperatures and precipitation. Furthermore, they affect a range of components of the Earth system as far to the south as the tropics. With both setups being similarly plausible, our findings highlight the need for more observational constraints on the permafrost hydrology to reduce the inter-model spread in Arctic climate projections.

Plant Hydraulics Improves Predictions of ET and GPP Responses to Drought

AbstractLand surface models typically ignore plant hydraulic processes and use empirical soil moisture stress functions to limit transpiration and photosynthesis. How plant hydraulics impacts the predictions of carbon and water fluxes and their responses to drought remains unclear. Here, we developed an analytical plant hydraulic scheme wherein hydraulic capacitances at the leaf level and stem level were considered, and incorporated it into the vegetation interface processes model (hereafter, VIP‐PHS). Then, we calibrated the VIP‐PHS parameters using a Markov Chain Monte Carlo (MCMC) method and conducted an ecosystem‐scale evaluation of the VIP‐PHS at 30 FLUXNET sites. The results showed that VIP‐PHS improved evapotranspiration (ET) and gross primary production (GPP) simulations across the studied sites, reducing the root mean square error (RMSE) by 0.02–0.27 mm day−1 and 0.04–0.52 g C m−2 day−1 in relation to the empirical soil hydraulic scheme (hereafter, VIP‐SHS). VIP‐PHS decreased the RMSE in ET and GPP by up to 45% and 55% in evergreen needleleaf forests. Compared to VIP‐SHS, VIP‐PHS decreased the sensitivities of ET and GPP to soil moisture stress, and significantly improved ET and GPP predictions under low soil moisture conditions. The sensitivity analysis showed that larger stem hydraulic capacitance better relieved xylem water stress, resulting in greater transpiration and GPP and less‐negative daily minimum stem water potential. These findings highlight the need to incorporate plant hydraulics into the next generation of Earth system models under future drought conditions.

The Role of Irrigation Expansion on Historical Climate Change: Insights From CMIP6

AbstractTo produce food for a growing world population, irrigated areas have increased from approximately 0.63 million km2 of land in 1900 to 3.1 million km2 of land in 2005. Despite this massive expansion, irrigation is still overlooked in most state‐of‐the‐art Earth system models (ESMs) involved in the Coupled Model Intercomparison Project phase 6 (CMIP6). To our knowledge, only three CMIP6 models represent irrigation activities: CESM2, GISS‐E2‐1‐G, and NorESM2‐LM. Here, we investigate the role of irrigation on historical climate at global and regional scales by analyzing trends of key surface climate variables in CMIP6 simulations during 1900–2014. The three models including irrigation show distinct behavior from the 15 models without irrigation over intensively irrigated areas (i.e., >50% of grid cell area is equipped by irrigation): both annual (months that correspond to monthly air temperature higher than 274 K) mean latent heat flux (LHF) and soil moisture increase over time, in contrast to the models without irrigation that show no trend or even a negative trend. The positive LHF trend over intensively irrigated areas in the irrigation‐on models is consistent with three satellite‐based LHF products. While annual (considering the warmest month in a year) warming trends are found in these regions for most of the no‐irrigation models, the increase in LHF induces a cooling trend for the models with irrigation, which was not confirmed by observational air temperature data sets. These findings, supported by satellite data, indicate the importance of improved representation of land management in the next generation of ESM.

Diazotrophy as a key driver of the response of marine net primary productivity to climate change

Abstract. The impact of anthropogenic climate change on marine net primary production (NPP) is a reason for concern because changing NPP will have widespread consequences for marine ecosystems and their associated services. Projections by the current generation of Earth system models have suggested decreases in global NPP in response to future climate change, albeit with very large uncertainties. Here, we make use of two versions of the Institut Pierre-Simon Laplace Climate Model (IPSL-CM) that simulate divergent NPP responses to similar high-emission scenarios in the 21st century and identify nitrogen fixation as the main driver of these divergent NPP responses. Differences in the way N fixation is parameterised in the marine biogeochemical component PISCES (Pelagic Interactions Scheme for Carbon and Ecosystem Studies) of the IPSL-CM versions lead to N-fixation rates that are either stable or double over the course of the 21st century, resulting in decreasing or increasing global NPP, respectively. An evaluation of these two model versions does not help constrain future NPP projection uncertainties. However, the use of a more comprehensive version of PISCES, with variable nitrogen-to-phosphorus ratios as well as a revised parameterisation of the temperature sensitivity of N fixation, suggests only moderate changes in globally averaged N fixation in the 21st century. This leads to decreasing global NPP, in line with the model-mean changes of a recent multi-model intercomparison. Lastly, despite contrasting trends in NPP, all our model versions simulate similar and significant reductions in planktonic biomass. This suggests that projected plankton biomass may be a more robust indicator than NPP of the potential impact of anthropogenic climate change on marine ecosystems across models.

Dynamics and characteristics of dry and moist heatwaves over East Asia

The increasing frequency of heatwaves over East Asia (EA) is impacting agriculture, water management, and people’s livelihood. However, the effect of humidity on high-temperature events has not yet been fully explored. Using observations and future climate change projections conducted with the latest generation of Earth System models, we examine the mechanisms of dry and moist heatwaves over EA. In the dry heatwave region, anticyclonic circulation has been amplified after the onset of heatwaves under the influence of the convergence of anomalous wave activity flux over northern EA, resulting in surface warming via adiabatic processes. In contrast, the moist heatwaves are triggered by the locally generated anticyclonic anomalies, with the surface warming amplified by cloud and water vapor feedback. Model simulations from phase six of the Coupled Model Intercomparison Project projected display intensification of dry heatwaves and increased moist heatwave days in response to projected increases in greenhouse gas concentrations.

Evaluation and optimisation of the I/O scalability for the next generation of Earth system models: IFS CY43R3 and XIOS 2.0 integration as a case study

Abstract. Earth system models have considerably increased their spatial resolution to solve more complex problems and achieve more realistic solutions. However, this generates an enormous amount of model data which requires proper management. Some Earth system models use inefficient sequential input/output (I/O) schemes that do not scale well when many parallel resources are used. In order to address this issue, the most commonly adopted approach is to use scalable parallel I/O solutions that offer both computational performance and efficiency. In this paper we analyse the I/O process of the European Centre for Medium-Range Weather Forecasts (ECMWF) operational Integrated Forecasting System (IFS) CY43R3. IFS can use two different output schemes: a parallel I/O server developed by Météo-France used operationally and an obsolete sequential I/O scheme. The latter is the only scheme that is being exposed by the OpenIFS variant of IFS. “Downstream” Earth system models that have adopted older versions of an IFS derivative as a component – such as the EC-Earth 3 climate model – also face a bottleneck due to the limited I/O capabilities and performance of the sequential output scheme. Moreover, it is often desirable to produce grid-point-space Network Common Data Format (NetCDF) files instead of the IFS native spectral and grid-point output fields in General Regularly-distributed Information in Binary form (GRIB), which requires the development of model-specific post-processing tools. We present the integration of the XML Input/Output Server (XIOS) 2.0 into IFS CY43R3. XIOS is an asynchronous Message Passing Interface (MPI) I/O server that offers features especially targeted at climate models: NetCDF output files, inline diagnostics, regridding, and, when properly configured, the capability to produce CMOR-compliant data. We therefore expect our work to reduce the computational cost of data-intensive (high-resolution) climate runs, thereby shortening the critical path of EC-Earth 4 experiments. The performance evaluation suggests that the use of XIOS 2.0 in IFS CY43R3 to output data achieves an adequate performance as well, outperforming the sequential I/O scheme. Furthermore, when we also take into account the post-processing task, which is needed to convert GRIB files to NetCDF files and also transform IFS spectral output fields to grid-point space, our integration not only surpasses the sequential output scheme but also the operational IFS I/O server.

A strong mitigation scenario maintains climate neutrality of northern peatlands

A strong mitigation scenario maintains climate neutrality of northern peatlands

Mesoscale Eddies Regulate Seasonal Iron Supply and Carbon Drawdown in the Drake Passage

AbstractThe Southern Ocean is a major region for uptake of anthropogenic carbon. The transport of carbon and nutrients are dominated by the Antarctic Circumpolar Current and its rich mesoscale eddy field; however, the current generation of Earth System Models (ESMs) lack the resolution to resolve eddies. Here we show that a computational model that explicitly represents mesoscale eddies can reproduce observed phases and amplitudes of seasonal biological productivity and partial pressure of carbon dioxide. Our sensitivity study demonstrates that when eddies are suppressed or parameterized, the model cannot reproduce these seasonal cycles. Experiments with suppressed eddy activity show that the lack of eddies significantly changes the iron supply and phenology of phytoplankton blooms. The mismatch in the timing and intensity of the bloom causes significant biases in the seasonal carbon cycle of the region with implications to the enigmatic biases in partial pressure of carbon dioxide among the state‐of‐the‐art ESMs.

Projected Land Evaporation and Its Response to Vegetation Greening Over China Under Multiple Scenarios in the CMIP6 Models

AbstractLand evaporation (ET) is of great significance in climate change research, water resource management, and numerical weather forecasting. In this study, Ridge Regression Method and Sensitivity Analysis Methods have been used to study the projected land evaporation changes over China, and its response to vegetation greening under low (Shared Socioeconomic Pathway [SSP]1‐2.6), medium (SSP2‐4.5), and high (SSP5‐8.5) forcing scenarios during 2020–2099, based on 16 of the latest generation of Earth System Models (ESMs) taking part in the Coupled Models Intercomparison Project Phase 6. Land evaporation is projected to significantly increase under all climate change scenarios, especially in the southern China where there is a humid climate. The influencing factors, including precipitation, air temperature, solar radiation, and leaf area index (LAI), are analyzed; LAI is indicated to dominate the interannual variations of land ET, contributing over 40% of the interannual variance in the warming climates. However, the sensitivity of land ET to vegetation greening is found to generally reduce along with the increasing radiation forcing levels. Such a reduced sensitivity is particularly true when making intermodel comparisons, possibly due to the uncertainties of vegetation parameterizations in different models. This study reveals the response of land ET to vegetation greening under multiple climate change scenarios, which may help to understand the essential role of vegetation in water cycle and provide support for future water resource management.

A Next Generation Ocean Carbon Isotope Model for Climate Studies I: Steady State Controls on Ocean 13C

AbstractThe 13C/12C of dissolved inorganic carbon (δ13CDIC) carries valuable information on ocean biological C‐cycling, air‐sea CO2 exchange, and circulation. Paleo‐reconstructions of oceanic 13C from sediment cores provide key insights into past as changes in these three drivers. As a step toward full inclusion of 13C in the next generation of Earth system models, we implemented 13C‐cycling in a 1° lateral resolution ocean‐ice‐biogeochemistry Geophysical Fluid Dynamics Laboratory (GFDL) model driven by Common Ocean Reference Experiment perpetual year forcing. The model improved the mean of modern δ13CDIC over coarser resolution GFDL‐model implementations, capturing the Southern Ocean decline in surface δ13CDIC that propagates to the deep sea via deep water formation. Controls on δ13CDIC of the deep‐sea are quantified using both observations and model output. The biological control is estimated from the relationship between deep‐sea Pacific δ13CDIC and phosphate (PO4). The δ13CDIC:PO4 slope from observations is revised to a value of 1.01 ± 0.02‰ (μmol kg−1)−1, consistent with a carbon to phosphate ratio of organic matter (C:Porg) of 124 ± 10. Model output yields a lower δ13CDIC:PO4 than observed due to too low C:Porg. The ocean circulation impacts deep modern δ13CDIC in two ways, via the relative proportion of Southern Ocean and North Atlantic deep water masses, and via the preindustrial δ13CDIC of these water mass endmembers. The δ13CDIC of the endmembers ventilating the deep sea are shown to be highly sensitive to the wind speed dependence of air‐sea CO2 gas exchange. Reducing the coefficient for air‐sea gas exchange following OMIP‐CMIP6 protocols improves significantly surface δ13CDIC relative to previous gas exchange parameterizations.

Improved representation of river runoff in Estimating the Circulation and Climate of the Ocean Version 4 (ECCOv4) simulations: implementation, evaluation, and impacts to coastal plume regions

Abstract. In this study, we improve the representation of global river runoff in the Estimating the Circulation and Climate of the Ocean Version 4 (ECCOv4) framework, allowing for a more realistic treatment of coastal plume dynamics. We use a suite of experiments to explore the sensitivity of coastal plume regions to runoff forcing, model grid resolution, and grid type. The results show that simulated sea surface salinity (SSS) is reduced as the model grid resolution increases. Compared to Soil Moisture Active Passive (SMAP) observations, simulated SSS is closest to SMAP when using daily, point-source runoff (DPR) and the intermediate-resolution LLC270 grid. The Willmott skill score, which quantifies agreement between models and SMAP, yields up to 0.92 for large rivers such as the Amazon. There was no major difference in SSS for tropical and temperate coastal rivers when the model grid type was changed from the ECCO v4 latitude–longitude–polar-cap grid to the ECCO2 cube–sphere grid. We also found that using DPR forcing and increasing model resolution from the coarse-resolution LLC90 grid to the intermediate-resolution LLC270 grid elevated the river plume area, volume, stabilized the stratification and shoal the mixed layer depth (MLD). Additionally, we find that the impacts of increasing model resolution from the intermediate-resolution LLC270 grid to the high-resolution LLC540 grid are regionally dependent. The Mississippi River Plume is more sensitive than other regions, possibly because the wider and shallower Texas–Louisiana shelf drives a stronger baroclinic effect, as well as relatively weak sub-grid vertical mixing and adjustment in this region. Since rivers deliver large amounts of freshwater and anthropogenic materials to coastal regions, improving the representation of river runoff in global, high-resolution models will advance studies of coastal hypoxia, carbon cycling, and regional weather and climate and will ultimately help to predict land–ocean–atmospheric feedbacks seamlessly in the next generation of Earth system models.

Projected Changes in Compound Flood Hazard From Riverine and Coastal Floods in Northwestern Europe

AbstractCompound flooding in coastal regions, that is, the simultaneous or successive occurrence of high sea levels and high river flows, is expected to increase in a warmer world. To date, however, there is no robust evidence on projected changes in compound flooding for northwestern Europe. We combine projected storm surges and river floods with probabilistic, localized relative sea‐level rise (SLR) scenarios to assess the future compound flood hazard over northwestern coastal Europe in the high (RCP8.5) emission scenario. We use high‐resolution, dynamically downscaled regional climate models (RCM) to drive a storm surge model and a hydrological model, and analyze the joint occurrence of high coastal water levels and associated river peaks in a multivariate copula‐based approach. The RCM‐forced multimodel mean reasonably represents the observed spatial pattern of the dependence strength between annual maxima surge and peak river discharge, although substantial discrepancies exist between observed and simulated dependence strength. All models overestimate the dependence strength, possibly due to limitations in model parameterizations. This bias affects compound flood hazard estimates and requires further investigation. While our results suggest decreasing compound flood hazard over the majority of sites by 2050s (2040–2069) compared to the reference period (1985–2005), an increase in projected compound flood hazard is limited to around 34% of the sites. Further, we show the substantial role of SLR, a driver of compound floods, which has frequently been neglected. Our findings highlight the need to be aware of the limitations of the current generation of Earth system models in simulating coastal compound floods.

Mid-Century Changes in the Mean and Extreme Climate in the Kingdom of Saudi Arabia and Implications for Water Harvesting and Climate Adaptation

The Kingdom of Saudi Arabia (KSA) is a water-scarce region with a dry, desert climate, yet flood-producing precipitation events and heat extremes lead to loss of life and damages to local infrastructure, property and economy. Due to its distinctive natural and man-made spatial features (e.g., coastal features, wadis, agricultural areas) studying changes in the mean climate and extreme events requires higher-resolution climate projections than those available from the current generation of Earth System Models. Here, a high-resolution convection-permitting regional climate model is used to downscale the middle of the 21st century (2041–2050) climate projections of the Community Earth System Model (CESM) under representative concentration pathway (RCP) 8.5 and for a historical time period (2008–2017) focusing on two months (August and November) within KSA’s dry-hot and wet seasons, where extreme events have historically been observed more frequently. Downscaling of climate reanalysis is also performed for the historical time period (2008–2017) to evaluate the downscaling methodology. An increase in the intensity and frequency of precipitation events is found in August by mid-century, particularly along the mountainous western coast of KSA, suggesting potential for water harvesting. Conversely, the northern flank of the Empty Quarter experiences a noticeable reduction in mean and extreme precipitation rates during the wet season. Increasing August heat index is found to particularly make regional habitability difficult in Jeddah by mid-century.

Reconciling Observation and Model Trends in North Atlantic Surface CO2

AbstractThe North Atlantic Ocean is a region of intense uptake of atmospheric CO2. To assess how this CO2 sink has evolved over recent decades, various approaches have been used to estimate basin‐wide uptake from the irregularly sampled in situ CO2 observations. Until now, the lack of robust uncertainties associated with observation‐based gap‐filling methods required to produce these estimates has limited the capacity to validate climate model simulated surface ocean CO2 concentrations. After robustly quantifying basin‐wide and annually varying interpolation uncertainties using both observational and model data, we show that the North Atlantic surface ocean fugacity of CO2 (fCO2−ocean) increased at a significantly slower rate than that simulated by the latest generation of Earth System Models during the period 1992–2014. We further show, with initialized model simulations, that the inability of these models to capture the observed trend in surface fCO2−ocean is primarily due to biases in the models' ocean biogeochemistry. Our results imply that current projections may underestimate the contribution of the North Atlantic to mitigating increasing future atmospheric CO2 concentrations.

Data-driven ENZYme (DENZY) model represents soil organic carbon dynamics in forests impacted by nitrogen deposition

Soil microorganisms participate in almost all soil organic carbon (SOC) transformations, but they are not represented explicitly in the current generation of earth system models. This study used a data-driven approach to incorporate extracellular enzyme activity into the Terrestrial ECOsystem (TECO) model, and the updated version was named the Data-driven ENZYme (DENZY) model. DENZY is based on results from an extensive data synthesis, which show that the CN ratio is positively correlated with ligninase activity (R2 = 0.50). The latter is inversely correlated to soil organic carbon storage. The DENZY model was parameterized using the revise database to information from a recent meta-analysis and tested for its ability to simulate SOC dynamics at Duke Forest (North Carolina, USA) from 1996 to 2007. DENZY can well simulate the observed negative relationship between ligninase activity and SOC under N deposition conditions (R2 ranges from 0.61 to 0.89). Moreover, outputs from DENZY better matched the observed SOC than its prototype model with the same parameterization. This study provides a simple and straightforward approach to effectively use real-world observations to improve SOC projections in terrestrial biogeochemical models.

Terrestrial ecosystem model performance in simulating productivity and its vulnerability to climate change in the northern permafrost region

AbstractRealistic projection of future climate‐carbon (C) cycle feedbacks requires better understanding and an improved representation of the C cycle in permafrost regions in the current generation of Earth system models. Here we evaluated 10 terrestrial ecosystem models for their estimates of net primary productivity (NPP) and responses to historical climate change in permafrost regions in the Northern Hemisphere. In comparison with the satellite estimate from the Moderate Resolution Imaging Spectroradiometer (MODIS; 246 ± 6 g C m−2 yr−1), most models produced higher NPP (309 ± 12 g C m−2 yr−1) over the permafrost region during 2000–2009. By comparing the simulated gross primary productivity (GPP) with a flux tower‐based database, we found that although mean GPP among the models was only overestimated by 10% over 1982–2009, there was a twofold discrepancy among models (380 to 800 g C m−2 yr−1), which mainly resulted from differences in simulated maximum monthly GPP (GPPmax). Most models overestimated C use efficiency (CUE) as compared to observations at both regional and site levels. Further analysis shows that model variability of GPP and CUE are nonlinearly correlated to variability in specific leaf area and the maximum rate of carboxylation by the enzyme Rubisco at 25°C (Vcmax_25), respectively. The models also varied in their sensitivities of NPP, GPP, and CUE to historical changes in climate and atmospheric CO2 concentration. These results indicate that model predictive ability of the C cycle in permafrost regions can be improved by better representation of the processes controlling CUE and GPPmax as well as their sensitivity to climate change.