Carbon dioxide removal (CDR) is proposed to limit the level of global warming and minimize the impacts of climate crises. However, how permafrost may respond to negative carbon emissions remains unknown. Here, the response of near-surface permafrost in the Northern Hemisphere is investigated based on idealized carbon dioxide (CO2) ramp-up (284.7-1138.8 ppm) and symmetric ramp-down model experiments. The results demonstrate that the timing of the minimum permafrost area lags the maximum CO2 concentration for decades, which is also observed in soil temperatures at different depths and active layer thicknesses (ALTs). When the CO2 concentration is reversed to the preindustrial level, the permafrost area decreases by similar to 12% relative to the initial conditions, together with additional warming in the ground temperature at the top of the permafrost, indicating the hysteresis of permafrost to CO2 removal. The most profound hysteretic responses occur at high latitudes for soil temperatures owing to Arctic amplification and at the southern margins of the permafrost zones for permafrost and ALT that largely linked to the climate state. Moreover, the sensitivity of permafrost and the associated thermodynamic factors to CO2 change is generally lower during the CO2 ramp-down phase than during the ramp-up phase, likely due to the release of stored heat on land. The results reveal the behaviour of permafrost in response to negative carbon emissions, which is informative for the projections of permafrost towards carbon neutral targets. In addition, the results may provide a reference for permafrost-related tipping points (e.g. releasing long-term stored greenhouse gases and destabilising recalcitrant soil carbon) and risk management in the future.

Remotely-sensed climate data records (CDRs) provide a basis for spatially distributed global climate model (GCM) inputs and validation methods. GCMs can take advantage of land surface models (LSMs), which aim to resolve surface energy, water and carbon budgets and hence these LSMs present important boundary conditions at the land-atmosphere interface. Pertinently, satellite data assimilation approaches are essential for improved land surface modelling for northern high latitudes ecosystems where permafrost degradation is reported to be ongoing. Permafrost, however, is an Essential Climate Variable (ECV) that cannot directly be monitored from space. Here, we advocate that CDRs, such as those compiled under the European Space Agency (ESA) Climate Change Initiative (CCI) programme, may be used in combination with permafrost models to improve our understanding of permafrost extent and degradation in a changing climate system. We describe the current types of remotely-sensed surface feature products that are widely used as indicators for permafrost related features. Furthermore, we highlight issues of using these site-specific permafrost proxies related to spatial scale, as well as the uncertainties in establishing present-day permafrost extent itself. Our assessment of the key ECVs that impact on permafrost, demonstrates how models that incorporate EO CDRs have the potential to boost our knowledge of permafrost conditions through better parametrisation of the thermal regime of permafrost soils. (C) 2017 The Authors. Published by Elsevier Inc.