Page | 1 esd-2020-37: “A new view of heat wave dynamics and predictability over the Eastern Mediterranean” by Assaf Hochman, Sebastian Scher, Julian Quinting, Joaquim Pinto and Gabriele Messori. Point by point response to Reviewer 1: Reviewer 1: In this paper, the authors employ an approach from dynamical systems theory to quantify the (intrinsic) predictability of atmospheric states based on reanalysis data during cold and hot extremes over the Eastern Mediterranean. This is complemented with GEFS reforecasts, which are used to infer forecast uncertainty, or practical predictability. While the distinction and investigation of practical and intrinsic predictability is not new (e.g. Melhauser & Zhang, 2012), I am not aware of any comparable publications in the context of heatwaves. In addition, a simple Lagrangian model is used to reveal the origin of near-surface air masses during hot and cold extreme events. The resulting paper is nicely structured, not too lengthy and certainly an interesting read. I only have two minor comments and a few additional comments, questions and suggestions, as the manuscript is well written and understandable. Response: Thank you for the positive feedback. We plan on addressing all of the Reviewer’s comments in the revised version of the manuscript as described below. Reviewer 1: In Melhauser & Zhang (2012), the classic Lorenz (1969) paper is cited multiple times; whereas intrinsic predictability is first defined as “the extent to which prediction is possible if an optimum procedure is used”, it is then also related to knowledge of the ‘atmospheric state’. Now, here, the authors do not cite any study when they claim that “As opposed to the practical predictability, the intrinsic predictability only depends on the characteristics of the atmosphere itself.” Even though I understand that authors might want to stick to historical definitions, to me, it makes only little sense to limit the forecasting system to the atmosphere. An increasing amount of evidence shows that the Earth’s surface does not only supply the atmosphere with heat and moisture, but, to some extent, also exerts control over it (e.g. Koster et al., 2010; Dirmeyer et al., 2018). Knowledge of the land (and ocean) surface state thus implies improved predictability up to sub-seasonal timescales (thanks to, e.g., soil memory). To be clear, I believe the focus on the atmosphere in this study makes sense, but I would still like to question why this ‘intrinsic’ predictability should be purely atmospheric by any means. To me, it seems like Lorenz (1969) emphasized the knowledge of all governing equations as well as observing the initial state, and I do not see why this would not include