date: Mon, 3 Dec 2007 08:40:01 +0800 from: subject: RE: CRUTEM data set to: Thanks Phil - much appreciated. Best Regards John John Church Development Leader Seria North Flank Tel : +673-3374745 E-mail : j.church@shell.com -----Original Message----- From: P.Jones@uea.ac.uk [mailto:P.Jones@uea.ac.uk] Sent: 30 November 2007 21:57 To: Church, John BSP-DPE/5 Subject: RE: CRUTEM data set John, Brohan, P., Kennedy, J., Harris, I., Tett, S.F.B. and Jones, P.D., 2006: Uncertainty estimates in regional and global observed temperature changes: a new dataset from 1850. J. Geophys. Res. 111, D12106, doi:10.1029/2005JD006548. The IPCC WG1 Report is on the Boulder web site of Wg1. You need Ch 3. Paper is too large to eamil - even if I was in Norwich. Can be got from AGU site - if you have a subscription. There are no corrections/adjustments for urbanization applied. We have removed a few stations (~50) which were badly affected. Most are in the USA. How we allow for urbanization is described in the paper. It is done by expanding the error range - one-sidedly. There are several papers in th IPCC WG1 Ch3 that show that urbanization isn't an issue. They are referred to. We are essentially following the conclusions of a paper from 1990. Papers by David Parker (2004,2006), Peterson (US work) and Jones et al. (1990). I am working on a paper looking for urbanization effects in China. If I plot eastern Chinese land temps vs HadSST2 for the seas just east/SE of China, they have hardly any difference in trends. This is despite some Chinese work showing there is a big effect. There are loads of site moves at Chinese sites, which you need to take into account. IN China also there are no rural sites - by western standards. Finally I wouldn't put any faith in talks you might have heard that say you can show urban trends by looking at NCEP data. Read this paper to find out why Simmons, A.J., P.D. Jones, V. da Costa Bechtold, A.C.M. Beljaars, P.W. Kållberg, S. Saarinen, S.M. Uppala, P. Viterbo and N. Wedi, 2004: Comparison of trends and low-frequency variability in CRU, ERA-40 and NCEP/NCAR analyses of surface air temperature. J. Geophys. Res., 109, D24115, doi:10.1029/2004JD006306. Again this is too large to email. I don't have it with me either. I may be able to squeeze this one into an email when back next week. Cheers Phil > Phil > Thanks for yur speedy reply. Much appreciated. Is it possible you can > give me the links to the references you describe below ? > > Also just to clarify - ther are NO CORRECTIONS applied to the land based > CRUTEM data set ? On any global locations ? So the question is : how do > you take account of urbanisation in these temperatures measurements ? > > Thanks > John > > > > John Church > Development Leader Seria North Flank > Tel : +673-3374745 > E-mail : j.church@shell.com > > > -----Original Message----- > From: P.Jones@uea.ac.uk [mailto:P.Jones@uea.ac.uk] > Sent: 29 November 2007 02:16 > To: Church, John BSP-DPE/5 > Subject: Re: CRUTEM data set > > > > John, > In Spain. The details should be in Brohan et al. (2006) > in JGR. > No corrections are applied. The error range (one sided) > gets enlarged by the amount that is given in the paper > and also in Ch 3 of AR4. Error range is just for the SH, NH > and global - it may be in locally at each grid box, but > it will be exactly the same. Nothing depends on location. > > Back in Norwich on Monday. > > Cheers > Phil > > >> Phil >> I was given your e-mail address byt someone in the UK Met office. I had >> a >> query about the corrections that are applied to the land based datasets >> to >> account for the 'urban heat island' effect..... >> >> Given that the dataset is global (various countries etc) I was curious >> as >> to how this is done, what are the error bars around the corrections. >> Are >> the corrections standard or is there varying error bars depending on >> where >> the measurement is taken. >> >> Thanks for any insights. >> >> Regards >> John Church >> >> >> >> John Church >> Development Leader Seria North Flank >> Tel : +673-3374745 >> E-mail : j.church@shell.com >> >> > > > >