Regional controls on blue-green water partitioning under climate change in CMIP6

Hydrological systems are undergoing rapid change under increasing greenhouse gas concentrations, yet substantial uncertainty remains regarding how precipitation is partitioned into blue (runoff) and green (transpiration) water flows across regions (blue–green water partitioning). Global assessments can reveal dominant large-scale signals, but they may mask regional differences in controlling mechanisms that are critical for climate-impact interpretation and climate-service applications. Here, we investigate regional controls on blue-green water partitioning using Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations.

We analyse blue-green water partitioning in historical simulations and two contrasting future scenarios (SSP1-2.6 and SSP3-7.0), quantified from monthly runoff, transpiration, and precipitation, and evaluated in multi-decadal means. To ensure interpretability at typical CMIP6 spatial resolution, we focus on selected large-scale IPCC AR6 reference regions and use multi-model ensemble statistics to characterise spread and agreement. We further assess differences across individual Earth system models to contextualise ensemble behaviour and sources of uncertainty.

To identify the main factors shaping regional changes in blue-green water partitioning, we apply statistical learning methods that relate partitioning changes to candidate controls representing precipitation characteristics, atmospheric demand, soil moisture conditions, and vegetation functioning. We use an ensemble-based framework with interpretability diagnostics to assess the relative importance of these controls across regions and scenarios. This regional perspective aims to complement global analyses by highlighting where and why controlling factors differ across hydroclimatic regimes, providing decision-relevant context for ecosystem-relevant green-water changes and runoff-relevant blue-water availability in a changing climate.