date: Tue Aug 28 15:09:23 2001 from: Tim Osborn subject: NERC fellowship to: drdendro@ldeo.columbia.edu Ed, I am applying for a 5-year fellowship from the UK Natural Environment Research Council (NERC). The topic of my fellowship is to be combining climate model and climate proxy data, but to provide it with a sharper focus I will concentrate on what some people call the "forward modelling" approach - whereby the output of the climate model is fed into a model of the environmental proxy and then that series is compared with the proxy series itself. Being a fellowship rather than a standard research grant application, I don't need to be fully precise in terms of what models and what proxies etc. I will use, but obviously want to make it clear that I have a reasonable plan for the 5 year period of the fellowship. Climate models: I'm hoping to use HadCM3 and ECHAM4/HOPE under natural and all forcings for the past few centuries - and maybe other climate model runs if there are appropriate ones. Proxies: Tree-rings is the one that is related to your work, and the reason for this e-mail. But other proxies will be sea-level, glaciers, tropical corals, documentary, ice cores and boreholes. On tree-rings, the idea is simple (though implementation may not be!). Take the simulated climate/weather time series from the climate model and input them into some tree-growth model, obtain some measure of tree-growth that the model produces and compare this with appropriate tree-ring chronologies. I'd like to be able to cover both density and width (or maybe their combination, density*width, though I believe that temporal variations in this will be dominated by temporal variations in width), for both drought & temperature sensitive regions. HadCM3 actually has a built in vegetation model (TRIFFID, I think it's called) that produces annual series of productivity that could be related to growth. That is easy to use, but there are various difficulties and the model is certainly not comprehensive. So in addition I want to say that I'll take the output (hopefully daily time series) and put it through other more specific tree-growth models, and use them for the comparison. Keith is away in France at the moment, but will hopefully provide some input when he gets back. But in addition, I was hoping I could name you as a collaborator in this (don't worry, nothing time consuming will be required from you), and also ask what your thoughts were about models of tree-growth that are suitable for the US drought-sensitive network. Are there any? Phil Jones told me about some (rather complex) model developed by Shaskin, but I wondered what advice you might be me about this aspect of the fellowship proposal. Any thoughts on difficulties that might arise would also be useful. Any help gratefully appreciated! Best regards Tim