Understanding low-frequency climate variability and abrupt change is essential for assessing Earth system sensitivity under changing external forcing. Past warm climates provide a unique opportunity to investigate such mechanisms because they involve large changes in orbital forcing, greenhouse gases, vegetation, ocean circulation, and cryosphere state that are not represented in the short instrumental record. In this project, we will use the Earth system model (ESM) EC-Earth to perform computationally demanding long transient and targeted palaeoclimate simulations addressing slow climate evolution, multicentennial variability, and abrupt climate–ecosystem transitions. The main component of the proposed work is the continuation of long transient simulations already underway. We are currently performing three Holocene transient simulations initiated at 9 ka and run to 4 ka, each starting from a different AMOC state. These ensemble members are designed to isolate the role of initial ocean circulation conditions in shaping low-frequency climate variability during the Holocene. We seek resources to extend these simulations to the pre-industrial period. This will create, to our knowledge, the first long transient Holocene ensemble performed with a high-resolution ESM including fully coupled atmosphere, ocean, sea ice, and dynamic vegetation. The simulations will provide a framework to investigate how orbitally forced climate evolution interacts with internal variability and feedbacks, and how such interactions may precondition abrupt regional changes, including hydroclimatic shifts relevant to past societal development and decline. A second major component is the continuation of our transient simulation of the Last Interglacial, initiated at 130 ka. About 5,000 simulation years have already been completed, and we propose to extend the run by another 5,000 years. This experiment is likewise among the first of its kind using a high-resolution ESM over such long timescales. It will allow us to compare low-frequency variability, ocean circulation, and climate stability between two warm intervals with different forcing histories, and to assess whether threshold-like behaviour emerges under gradually evolving boundary conditions. These simulations directly support ongoing and planned research activities. They underpin our current VR-funded work on multicentennial climate variability, the submitted VR proposal on climate variability suppression and abrupt climate–ecosystem transitions, and the planned ERC Advanced Grant proposal on hydroclimate vulnerability in the Earth system. The outputs will be used to study mechanisms linking slow forcing, climate variability, ocean–atmosphere coupling, vegetation feedbacks, and abrupt transitions. In addition, we request resources to contribute to the Assessment Fast Track of CMIP7 within PMIP using the new model version EC-Earth4. Specifically, we will perform the abrupt-127k experiment following the CMIP7/PMIP7 protocol. These simulations will benchmark EC-Earth4 in a palaeoclimate context and ensure that our group contributes to CMIP7 and IPCC AR7. Because these simulations are computationally intensive and generate very large datasets, both compute and storage resources are essential. The requested allocation will partly enable completion of these simulations, participation in CMIP7, and delivery of data and analyses. We also apply other HPC resources such as ECMWF computing and archive facilities.