cc: "Phil Jones" , "Ag Stephens" , "Anna Steynor" , "Chris Kilsby" , "Geoff Jenkins" , "Mark Elkington" , "Richard Westaway" , "Hardy, Karl \(FM\)" , "Kay Jenkinson" , "Richard Lamb" date: Wed, 5 Dec 2007 17:06:54 -0000 from: "Roger Street" subject: Re: UKCIP08 core brief to: "David Sexton" , "Humphrey, Kathryn \(GA\)" In the context of the core briefing, although it might be closed off, I offer the following comments that may inform its future developments: Under Key Messages Second message, last point - I would suggest "A Weather Generator tool will be available to produce daily and hourly data consistent with the probabilistic projections over land". Message identified as 6, third point - I understand that sub-surface projections will be available at grid-square scale (and may or may not be available at an aggregated level). Under Q&A Under What does "probablistic" mean in the context of UKCIP08 o Agree with David regarding not using absolute frequency. I would suggest - "Probablistic" in this case means something different from the probability you get from repeatedly rolling a set of dice for example, as the UKCIP08 probabilities are a measure..." The key difference for me is that probabilities asssociated with rolling a set of dice are based on frequency of occurrence whereas those within UKCIP08 are based on projections of occurrence as derived through the particular set of assumptions and procedures that are at the basis of UKCIP08. Under "What does UKCIP08 give me that UKCIP02 didn't? o Where there is reference to the time-slices, I would suggest "...(UKCIP02 only used three sequential time slices to cover the century)" In terms of your request for information on uses of UKCIP08, some which you may also want to consider include the following: Headline messages and maps are often used to demonstrate to board members and owners the need for more detailed consideration, as a basis for further work (e.g., when working with stakeholders and partners). These can also be used for communications to taxpayers, stakeholders and shareholders. Interpretatation as to behavioral changes being undertaken or proposed when talking to 'customers' (e.g., changes in level of services or products, withdrawal or change in availability of service or product) Customisable output o decisions related to investments (e.g., irrigation systems; changes in land-use, cropping or plant species; changes in budget to support maintenance activities. water company infrastructure) o more detailed validation of recommendations and decisions to regulators and shareholders (e.g., backing up recommendations related to proposed investments or changes in investment strategies). o basis for revised standards and regulations (e.g., built environment and concerns related to overheating). o basis for decisions related to spacial planning and flood management Roger ----- Original Message ----- From: "David Sexton" To: "Humphrey, Kathryn (GA)" Cc: "Phil Jones" ; "Ag Stephens" ; "Anna Steynor" ; "Chris Kilsby" ; "Geoff Jenkins" ; "Mark Elkington" ; "Richard Westaway" ; "Roger Street" ; "Hardy, Karl (FM)" ; "Kay Jenkinson" ; "Richard Lamb" Sent: Tuesday, December 04, 2007 3:15 PM Subject: RE: UKCIP08 core brief > Hi, > > sorry to be nit-picky but both types of probability are absolute. > Instead of "absolute probability" how about "frequency-based > probability" or just "probability" because the rest of the sentence > defines it. > > Cheers, David > > On Tue, 2007-12-04 at 11:21 +0000, Humphrey, Kathryn (CEOSA) wrote: >> Hi all, >> >> Here's an updated version of the core brief based on your comments. Let >> me know if I've missed anything! (ignore spelling, grammar and numbering >> errors). >> >> Kathryn >> >> >> >> -----Original Message----- >> From: Phil Jones [mailto:p.jones@uea.ac.uk] >> Sent: 04 December 2007 10:31 >> To: Humphrey, Kathryn (CEOSA); David Sexton >> Cc: Ag Stephens; Anna Steynor; Chris Kilsby; Geoff Jenkins; Mark >> Elkington; Richard Westaway; Roger Street; Hardy, Karl (FM); Kay >> Jenkinson; Richard Lamb >> Subject: RE: UKCIP08 core brief >> >> >> Kathryn, >> No, it isn't! The first 2 sentences are OK, as is the last. >> The third is the crucial one. >> >> What you have is >> >> For UKCIP08, weather for future time periods is generated based on the >> probabilistic projections, so that the statistical properties are the >> same. >> >> What it needs to say is >> >> For UKCIP08, weather for future periods is generated based on >> perturbing the WG according to the probabilistic projections. >> >> A word like perturbing has to be there. The statistical >> properties of the WG get perturbed - based on the change >> factors from the probabilistic work. >> >> Can what I sent yesterday go in full into the glossary for WGs? >> >> I think the PMG will now be happy with this. >> >> Cheers >> Phil >> >> >> >> >> >> At 17:16 03/12/2007, Humphrey, Kathryn (CEOSA) wrote: >> >Thanks Phil. I've simplified this a bit for the core brief as if I >> give >> >"perturbations" and "trained" to a Minister they're not going to like >> >it! Is this still correct? >> > >> >Weather Generators (WGs) traditionally use existing weather data and >> >random number sampling to produce longer series of synthetic daily and >> >hourly weather data. The synthetic sequences will have similar >> >statistical properties as the observed data on which they are trained. >> >For UKCIP08, weather for future time periods is generated based on the >> >probabilistic projections, so that the statistical properties are the >> >same. The weather variables generated are: precipitation, mean >> >temperature, diurnal temperature range, sunshine duration, vapour >> >pressure and wind speed, as well as derived (from the above variables) >> >estimates of potential evapotranspiration and direct and diffuse >> >radiation. >> > >> > >> >-----Original Message----- >> >From: Phil Jones [mailto:p.jones@uea.ac.uk] >> >Sent: 03 December 2007 16:36 >> >To: David Sexton >> >Cc: Humphrey, Kathryn (CEOSA); Ag Stephens; Anna Steynor; Chris Kilsby; >> >Geoff Jenkins; Mark Elkington; Richard Westaway; Roger Street; Hardy, >> >Karl (FM); Kay Jenkinson; Richard Lamb >> >Subject: Re: UKCIP08 core brief >> > >> > >> > David et al, >> > I thought this bit of text had almost been signed off !! Anyway >> >you raise a good point. >> > >> > So how's about this - for the glossary, and for Kathryn to add the >> >UKCIP08 core brief. >> > >> > Weather Generator: Weather Generators (WGs) traditionally use >> >existing >> > weather data and random number sampling to produce longer series of >> >synthetic daily >> > and hourly weather data. The synthetic sequences will have similar >> >statistical >> > properties as the observed data on which they are trained. For >> >UKCIP08, weather for future time >> > periods is generated based on perturbations to the precipitation >> >and temperature >> > parameters of the WG where the perturbations are sampled from >> >probabilistic projections >> > of these key driving parameters . The weather variables generated >> >are: precipitation, >> > mean temperature, diurnal temperature range, sunshine duration, >> >vapour pressure and >> > wind speed, as well as derived (from the above variables) estimates >> >of potential evapotranspiration >> > and direct and diffuse radiation. >> > >> > This is essentially David's option 1, but I've been much more >> >specific as to which >> > parameters are perturbed in the WG. >> > >> > Cheers >> > Phil >> > >> > >> > >> > >> > >> >At 14:20 03/12/2007, David Sexton wrote: >> > >Sorry to be pain. Can I propose a tweak to the following wording. >> > > >> > >On Mon, 2007-12-03 at 13:25 +0000, Phil Jones wrote: >> > > > Weather Generator: Weather Generators (WGs) traditionally use >> > > > existing >> > > > weather data and random number sampling to produce longer series >> of >> > > > synthetic daily >> > > > and hourly weather data. The synthetic sequences will have >> similar >> > > > statistical >> > > > properties as the actual data on which they are trained. >> > > >> > > >> > > >> > > > For UKCIP08, weather for future >> > > > time periods is generated based on perturbations to the internal >> > > > parameters of the WG >> > > > derived from changes in the probabilistic daily PDFs. >> > > >> > >I don't really understand last part of above sentence, partly because >> I >> > >think daily PDFs are PDFs for each day, and using both probabilistic >> >and >> > >PDFs is a tautology, and I am not sure what changes in PDFs means, >> and >> >I >> > >am not sure if perturbations refers to change factors or different >> >model >> > >variants. So how about... >> > > >> > >For UKCIP08, weather for future time periods is generated based on >> > >perturbations to the driving parameters of the WG where the >> > >perturbations are sampled from probabilistic projections of those >> > >driving parameters. >> > > >> > >OR >> > > >> > >For UKCIP08, weather for future time periods is generated by first >> > >calculating probabilistic projections of the driving parameters of >> the >> > >WG and then sampling these several times and using them to drive >> >several >> > >WG series. >> > > >> > > >> > >I admit it is tricky but hopefully you can see what I am getting at. >> > > >> > >Cheers, David >> > > >> > > >> > > >> > > >> > > > The weather variables generated are: >> > > > precipitation, mean temperature, diurnal temperature range, >> >sunshine >> > > > duration, vapour pressure >> > > > and wind speed, as well as derived (from the above variables) >> > > > estimates of potential >> > > > evapotranspiration and direct and diffuse radiation. >> > >-- >> > >David Sexton PhD Climate Research Scientist >> > >Met Office Hadley Centre FitzRoy Road Exeter EX1 3PB UK >> > >Tel: +44 (0)1392 886524 Fax: +44 (0)1392 885681 >> > >E-mail: david.sexton@metoffice.gov.uk http://www.metoffice.gov.uk >> > >> >Prof. Phil Jones >> >Climatic Research Unit Telephone +44 (0) 1603 592090 >> >School of Environmental Sciences Fax +44 (0) 1603 507784 >> >University of East Anglia >> >Norwich Email p.jones@uea.ac.uk >> >NR4 7TJ >> >UK >> >----------------------------------------------------------------------- >> - >> >---- >> >Department for Environment, Food and Rural Affairs (Defra) >> > >> >This email and any attachments is intended for the named recipient >> only. >> >If you have received it in error you have no authority to use, >> disclose, >> >store or copy any of its contents and you should destroy it and inform >> >the sender. >> >Whilst this email and associated attachments will have been checked >> >for known viruses whilst within Defra systems we can accept no >> >responsibility once it has left our systems. >> >Communications on Defra's computer systems may be monitored and/or >> >recorded to secure the effective operation of the system and for other >> >lawful purposes. >> >> Prof. Phil Jones >> Climatic Research Unit Telephone +44 (0) 1603 592090 >> School of Environmental Sciences Fax +44 (0) 1603 507784 >> University of East Anglia >> Norwich Email p.jones@uea.ac.uk >> NR4 7TJ >> UK >> ------------------------------------------------------------------------ >> ---- >> > -- > David Sexton PhD Climate Research Scientist > Met Office Hadley Centre FitzRoy Road Exeter EX1 3PB UK > Tel: +44 (0)1392 886524 Fax: +44 (0)1392 885681 > E-mail: david.sexton@metoffice.gov.uk http://www.metoffice.gov.uk >