date: Wed, 20 May 2009 13:01:49 +0100 from: Clare Goodess subject: Fwd: Re: CIRCE urban heat island simulations for Athens to: p.jones@uea.ac.uk,c.harpham@uea.ac.uk >X-Authentication-Warning: ueamailgate01.uea.ac.uk: defang set sender >to using -f >X-IronPort-AV: E=Sophos;i="4.41,215,1241395200"; > d="txt'?scan'208";a="6133780" >Subject: Re: CIRCE urban heat island simulations for Athens >From: Mark McCarthy >To: CHRISTOS GIANNAKOPOULOS >Cc: Clare Goodess , maureen agnew >Date: Tue, 19 May 2009 12:16:32 +0100 >X-Mailer: Evolution 2.0.2 (2.0.2-35.0.4.el4_6.1) >X-Canit-CHI2: 0.00 >X-Bayes-Prob: 0.0001 (Score 0, tokens from: @@RPTN, f034) >X-Spam-Score: 0.00 () [Tag at 5.00] SPF(none,0) >X-CanItPRO-Stream: UEA:f034 (inherits from >UEA:10_Tag_Only,UEA:default,base:default) >X-Canit-Stats-ID: 22421490 - 429395a988f2 >X-Antispam-Training-Forget: >https://canit.uea.ac.uk/b.php?i=22421490&m=429395a988f2&c=f >X-Antispam-Training-Nonspam: >https://canit.uea.ac.uk/b.php?i=22421490&m=429395a988f2&c=n >X-Antispam-Training-Spam: >https://canit.uea.ac.uk/b.php?i=22421490&m=429395a988f2&c=s >X-Scanned-By: CanIt (www . roaringpenguin . com) on 139.222.131.184 > >Hi Christos, > > Thanks for this. Glad you downloaded the data OK. Answers to your >questions are: > >1) In section 3.1 of my report I use the urban tile from NoUrbNoAnth and >define the urban heat island as the difference between the urban tile >and the gridcell mean. In this particular experiment the urban coverage >is set to zero so that there is no feedback of the urban surface onto >the gridcell mean. This was done for simplicity. For section 3.2 I also >reference the other experiments, but I do use the urban tile >temperatures. > >2) Yes the missing gridcells are sea points in the model. > >3) Model resolution is 25km. It is on a rotated pole grid so the cells >are not aligned N-S. I attach a file of lat-lon coords for the corners >of the gridcells. You will also notice that it does not fit neatly over >Athens. The model will have the peak heat island where there is peak >urban coverage. Therefore gridcell 5 is defined as the "Athens" gridcell >for the model because it has the highest urban fraction (40%). You >should already have the fractional coverage of surface types for each >gridcell but let me know if you do not. > >4) No I do not use neighbouring grid cells. There are significant >temperature differences between the "central" athens gridcell and its >neighbours even without the inclusion of urban areas. Therefore I >defined the urban heat island using the NoUrbNoAnth experiment (see >answer to quest. 1 above). This will be an underestimate because we do >not account for the feedback into the atmosphere. Defining what the >urban heat island actually is for both model and observations is rather >tricky. > >With reference to your plots: > >Fig.1 What are the observations? Are they a single station or from a >gridded product? > >The model appears to have a significant summer warm bias. For such a >local comparison we shouldn't expect to achieve quantitative agreement >between the model and the observations. The purpose of the frequency of >hot nights/days in the report is to show the sensitivity of extremes to >urbanisation and climate change. You could always try applying a simple >bias correction to the model temperatures and then recalculate the >exceedences to see if that brings the model and observations more in- >line. > >Kind regards, >Mark > >On Fri, 2009-05-15 at 13:12 +0300, CHRISTOS GIANNAKOPOULOS wrote: > > Dear Mark, > > > > we have started to process the UHI data for Athens from the data we have > > dowloaded from the website you have pointed out. > > Our first task was to try to reproduce the figures in your CIRCE UHI > > report. The attached file contans some of our graphs where we have also > > added the observations from the station of NOA in the centre of the city. > > Some questions remain: > > 1) we believe that in your figure reports you are using the > temperature from > > the tiles and not the gridcell mean. Pls confirm > > 2) for Athens some gridcells are missing (7, 8,9) from the files we have > > downloaded. We believe they are missing becasue they are sea gridcells. Pls > > confirm. > > 3) Can you provide us the exact coordinates of the cells for the files in > > the website? What is the model resolution (25km?)? > > 4) In your report to calcuate the UHI, you use all the neighboring cells of > > the city? > > > > Please let me know if you want more clarifications on these. > > We will be happy to provide more plots if needed. > > Best, > > Christos > > > > Dr. Christos Giannakopoulos > > Senior Researcher > > Institute for Environmental Research and Sustainable Development > > National Observatory of Athens > > V.Pavlou and Metaxa Str., P.Pendeli > > GR-15236 Athens, Greece > > Tel: +30-210-8109128 > > Fax: +30-2108103236 > > email: cgiannak@meteo.noa.gr >-- >Mark McCarthy Climate Impacts Scientist >Met Office Hadley Centre, Fitzroy Road, Exeter, EX1 3PB >Tel: +44(0)1392 884672 Fax: +44(0)1392 885681 >E-mail: mark.mccarthy@metoffice.gov.uk http://www.metoffice.gov.uk > Dr Clare Goodess Climatic Research Unit School of Environmental Sciences University of East Anglia Norwich NR4 7TJ UK Tel: +44 -1603 592875 Fax: +44 -1603 507784 Web: http://www.cru.uea.ac.uk/ http://www.cru.uea.ac.uk/~clareg/clare.htm Attachment Converted: "c:\eudora\attach\athenscoords.txt"