cc: rgjones@meto.gov.uk (Richard Jones) date: Sun, 25 Apr 1999 22:52:03 +0100 (BST) from: Richard Jones subject: Comments on 10.1 to: tar10@egs.uct.ac.za (TAR 10 mailbox) Dear All, Here are my first thoughts on the draft of 10.1 - constructive comments I hope which I am prepared to have shot down! 10.1 para 1: Another reason for the high level of uncertainty surely is simply that the resolution of models to date have precluded the direct extraction of useful information at fine scales simply because they have no skill at their gridscale or below. Also, the last sentence implies regional climate is determined by processes that span many scales and are non-linear hence is poorly modelled and so global climate, which we can model reasonably well, is not determined by such processes - though in reality it is. I think I know what you are trying to say but the statement is too strong (and misses important points about the direct effects of lack of resolution as I have implied above). 10.1 para 4 sentence 5. I would replace beginning by starting. 10.1.1 para 1 sentence 4: I would rather start with "Biases in the simulation of present day climate compared ..." 10.1.2 para 1: I think the lower partition of the scale ranges should be 10^4 km^2 as this corresponds to 2x2 gridboxes of a 50km model and hence about the limit of skill of most current RCMs. I know this is somewhat arbitrary, but it does give a sensible reference point - if 10^3 km^2 is used then we should give a reason, a typical length-scale of some important local forcing or something similar. 10.1.2 para 3 sentence 3: I think "These interactions ... can be highly non-linear" is more accurate. I think if we push this non-linearity idea too hard we are in danger of implying the problem is intractible without very high resolution GCMs with greatly improved physics. Best wishes, Richard. -- Richard Jones (Rm H304) - Tel. (44) 01344 856418 Fax. (44) 01344 854898 Hadley Centre for Climate Prediction and Research, UK Met. Office, London Road, Bracknell, RG12 2SZ, UK. Email - local : hadrj@hc0500 : - Internet : rgjones@meto.gov.uk