Abstract:This study compares two ensemble methods, pseudo global warming (PGW) and multimodel ensemble (MME) mean in evaluating future vegetation‐climate changes and feedback over East Asia using a regional model driven with boundary conditions derived from historical and RCP8.5 runs of four global models. Over most of the domain, MME and PGW produce similar future climate changes, except for two regions: over Northeast Asia, MME projects less winter warming and less increase of summer precipitation than PGW, and over North China Plain, MME projects greater increase of summer precipitation than PGW. Both PGW and MME project greater leaf area index (LAI) and expansion of tree coverage at the expense of grass over most of the domain due to elevated CO2 concentration and climate change. However, over Northeast China, PGW and three ensemble members project an increase of temperate trees at the expense of boreal trees or grass and a decrease of LAI, and one ensemble member projects a loss of forest to grassland. Both PGW and MME identify two hotspots of vegetation‐temperature feedback, including the Northeast China where the projected loss of boreal evergreen forest causes a significant winter cooling (which offsets a major portion of greenhouse gas‐induced warming) and the Tibetan Plateau where the projected LAI increase and vegetation expansion lead to a significant winter warming through reduced albedo. The vegetation feedback impact on precipitation is similar between the two approaches, including a significant wet signal in a belt between 30°N and 40°N during winter and over the Tibetan Plateau during summer. Climate is undergoing severe changes. Climate simulation and projections from global climate models suffer from coarse resolution and model dependence, especially for the East Asia region. In this study, we assess multimodel ensemble (MME) projections of the East Asian vegetation‐climate changes using a regional model driven with boundary conditions from multiple global models. Two ensemble approaches are used, the traditional MME mean that is based on unweighted average of the four runs driven with individual global models and the reanalysis‐based pseudo global warming (PGW) that derives the regional model's boundary conditions by adding the four‐model average of future changes to the boundary conditions from the reanalysis data. PGW and MME produce a similar future climate change signals over most of the model domain except for the Northeast China, North China, and India regions. Both PGW and MME identify two hotspots of vegetation‐temperature feedback, including the Northeast China with significant winter cooling due to forest loss and the Tibetan Plateau with significant winter warming due to LAI increase and vegetation cover expansion. Two ensemble approaches are used to project the future vegetation‐climate changes over East Asia Projected changes over Northeast China are highly sensitive to the GCM‐derived boundary conditions Vegetation feedback has significant impacts on winter temperature over Northeast China and Tibetan Plateau due to albedo effects

Optimization of Multi-Ecosystem Model Ensembles to Simulate Vegetation Growth at the Global Scale

Pixel-level parameter optimization of a terrestrial biosphere model for improving estimation of carbon fluxes with an efficient model–data fusion method and satellite-derived LAI and GPP data

Spatial parameter optimization of a terrestrial biosphere model for improving estimation of carbon fluxes for deciduous forests in the eastern United States: an efficient model-data fusion method

Photosynthetic Productivity and Its Efficiencies in ISIMIP2a Biome Models: Benchmarking for Impact Assessment Studies

Evaluating the performance of CMIP6 Earth system models in simulating global vegetation structure and distribution

On the Use of Machine Learning Based Ensemble Approaches to Improve Evapotranspiration Estimates from Croplands Across a Wide Environmental Gradient

Comparison of Satellite-Based Models for Estimating Gross Primary Productivity in Agroecosystems

Intercomparison of global terrestrial carbon fluxes estimated by MODIS and Earth system models

Multimodel Future Projections of the Regional Vegetation‐Climate System Over East Asia: Comparison Between Two Ensemble Approaches

Optimization of ecosystem model parameters through assimilating eddy covariance flux data with an ensemble Kalman filter

Fusion of multiple models for improving gross primary production estimation with eddy covariance data based on machine learning

Evaluation of Crop Model Prediction and Uncertainty Using Bayesian Parameter Estimation and Bayesian Model Averaging

Optimization of ecosystem model parameters using spatio-temporal soil moisture information

Multi-model ensemble prediction of terrestrial evapotranspiration across north China using Bayesian model averaging

A Multiple Crop Model Ensemble for Improving Broad-Scale Yield Prediction Using Bayesian Model Averaging

Application of the Ecosystem Model and Markov Chain Monte Carlo for Parameter Estimation and Productivity Prediction

The Optimization of Model Ensemble Composition and Size Can Enhance the Robustness of Crop Yield Projections

Simulating canopy carbonyl sulfide uptake of two forest stands through an improved ecosystem model and parameter optimization using an ensemble Kalman filter

Toward “optimal” Integration of Terrestrial Biosphere Models

Multi-model ensembles for ecosystem prediction

Comparing the Impacts of Single- and Multi-Objective Optimization on the Parameter Estimation and Performance of a Land Surface Model