A theoretical index for understanding distinct land relative humidity trends in observations, reanalyses, and models

Key findings from this study

• The study found that an index based on precipitation to modified potential evapotranspiration captures relative humidity variability and discrepancies among observations, reanalyses, and models.
• The authors report that nearly all Earth system models underrepresent historical land relative humidity decrease, with most models simulating weaker drying than observations indicate.
• The researchers demonstrate that weaker model relative humidity decreases originate primarily from insufficient subtropical precipitation decline linked to muted subtropical high intensification.

Overview

A simple index based on the ratio of precipitation to modified potential evapotranspiration captures spatiotemporal variability in land surface relative humidity and reconciles discrepancies among observations, reanalyses, and Earth system models. The framework enables physical calibration of biased reanalyses and quantitative interpretation of relative humidity changes across multiple forcing factors. Over 1973–2024, land relative humidity decreased substantially due to rising potential evapotranspiration with temperature and stagnant precipitation increases.

Methods and approach

The authors developed an index defined by the ratio of precipitation to a modified potential evapotranspiration formulated independently of relative humidity. This index was evaluated against observed relative humidity, reanalysis data, and Earth system model simulations spanning 1973–2024. The index characterizes spatiotemporal variability and quantifies divergence between different datasets. Model-observation discrepancies were attributed to precipitation and potential evapotranspiration contributions using the index framework.

Results

Reanalyses overestimate observed relative humidity decreases, consistent with exaggerated surface warming and precipitation decline. The index captures this coherent bias and enables calibration using observed precipitation and temperature data. Earth system models simulate a wide range of relative humidity trends, with nearly all underrepresenting historical drying. Weaker model simulations of relative humidity decrease arise primarily from insufficient subtropical precipitation decline, linked to muted intensification of subtropical highs and biased subtropical climatology. Model-observation discrepancy cannot be explained by internal variability alone, indicating systematic model underestimation of forced relative humidity decrease.

Implications

The proposed index provides a diagnostic framework for evaluating relative humidity representation across observational networks, reanalysis products, and climate models. By decomposing relative humidity changes into precipitation and potential evapotranspiration components, the index enables identification of specific sources of model bias and discrepancy. This capability strengthens confidence in attributing observed relative humidity trends to distinct physical mechanisms rather than measurement artifacts or data processing artifacts. The framework facilitates improved reanalysis calibration and more accurate constraint of model behavior against observations, advancing understanding of coupled land-atmosphere processes.

The finding that models systematically underestimate historical relative humidity drying has direct implications for projections of future hydroclimate and terrestrial ecosystem resilience. If models underrepresent forced responses in relative humidity, current climate projections may underestimate drying stress on vegetation and agricultural systems. The index framework enables more rigorous evaluation of model skill in simulating regional precipitation and potential evapotranspiration changes, identifying which models warrant greater weight in multi-model ensemble projections. Enhanced understanding of model limitations supports development of constrained future projections with lower uncertainty bounds.

The discrepancy between models and observations implicates specific regional processes, particularly subtropical high-pressure system intensification and associated precipitation responses. Future model development should prioritize improved representation of subtropical atmospheric circulation changes and their interaction with land surface processes. The index provides a quantitative metric for assessing whether model refinements successfully reduce the observed-model gap in relative humidity trends, supporting iterative model improvement toward more skillful hydroclimate prediction.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.