A Bias‐Robust Framework for Quantifying Community Responses to the Climate Change Using the Occurrence Data

Aim Assessing species range shifts and subsequent thermophilization—the increasing dominance of warm‐adapted species in communities—is critical for understanding ecological responses to climate change. Because systematic monitoring data for assessing thermophilization are limited geographically, temporally, and taxonomically, the use of occurrence data has been attempted. Although occurrence data have enabled important discoveries in studies evaluating the impact of climate change‐induced range shifts on community responses, two major biases—Spatiotemporal Variation of Observation Effort (STVOE) and the Truncation Effect—hinder their full potential. This study introduces the Community Change Detection Model (CCDM), a novel statistical framework designed to address these biases and accurately measure the thermophilization rate using large‐scale occurrence data. Innovation CCDM is a simple method based on multiple regression, yet it is the first method to simultaneously correct for the various degrees of STVOE and the truncation effect, enabling reliable thermophilization rate estimation from biased occurrence data. Validated through simulations and an empirical application to Japanese vascular plants, CCDM surpasses previous approaches, enabling the use of large amounts of occurrence data that had previously been restricted. Main Conclusions Applying CCDM to Japan's vascular plant communities from 1900 to 2024 revealed consistent thermophilization with acceleration since 1985, though it lagged behind regional warming rates, leading to the development of a climatic debt. Our analysis enabled this discovery by utilising occurrence data from historical warming periods, which is largely inaccessible through systematic monitoring. Furthermore, CCDM demonstrated robust performance in estimating thermophilization rates under various bias conditions in the simulation, indicating its high general applicability. CCDM unlocks the potential of historical occurrence data, offering unprecedented insights into climate change impacts across under‐monitored regions, periods, and taxa.