English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multi-model ensemble analysis integrates information from multiple climatemodels into a unified projection. However, existing integration approachesbased on model averaging can dilute fine-scale spatial information and incurbias from rescaling low-resolution climate models. We propose a statisticalapproach, called NN-GPR, using Gaussian process regression (GPR) with aninfinitely wide deep neural network based covariance function. NN-GPR requiresno assumptions about the relationships between models, no interpolation to acommon grid, no stationarity assumptions, and automatically downscales as partof its prediction algorithm. Model experiments show that NN-GPR can be highlyskillful at surface temperature and precipitation forecasting by preservinggeospatial signals at multiple scales and capturing inter-annual variability.Our projections particularly show improved accuracy and uncertaintyquantification skill in regions of high variability, which allows us to cheaplyassess tail behavior at a 0.44$^\circ$/50 km spatial resolution without aregional climate model (RCM). Evaluations on reanalysis data and SSP245 forcedclimate models show that NN-GPR produces similar, overall climatologies to themodel ensemble while better capturing fine scale spatial patterns. Finally, wecompare NN-GPR's regional predictions against two RCMs and show that NN-GPR canrival the performance of RCMs using only global model data as input.

Multi-model Ensemble Analysis with Neural Network Gaussian Processes