FutureClim Update: 2080s added to the mix!
Koo, Jawoo (IFPRI)
Hi CSI’ers!
Today, FutureClim added a new time-slice (2080s) to the mix - now the database includes the total of 37 (baseline 2000 plus (4 GCM’s x 3 SRES’ x 3 time-slices)) global 5 arc-minute datasets with monthly climate variables (including solar radiation, tmax, tmin, rainfall, and rainy days).
If you have downloaded the dataset before, you should have received the update notice by now (let me know if you haven’t). Otherwise, please visit http://futureclim.info for data download instructions.
As always, if you have *any* question on the correct use/interpretation of the dataset, please don’t hesitate to contact one of the authors (copied).
Cheers!
Jawoo
Paul Georgie
Dear CSI’rs,
For the last 11 months I have been adding to my knowledge of development studies and experience to that of my geographic heart and passion for information systems. My studies here at Edinburgh University’s GeoSciences Department has created an opportunity to utilise agent based modelling for the creation of a new food security model that is in parallel with the ever-changing minds of the development game, now represented by the ‘new’ four pillars of food security.
In order to develop the model upon a comparative scale of data availability to the worsening gaps that exist in the majority world, the pilot is based upon a low and middle-income country. Juxtaposing the pillars and the predictions of longer-term impacts of climate change, both human and environmental, I have developed this model, albeit simple. However, my question lies as to how a model such as this could scale down such climate data and be able to utilise competent climatic algorithms (if they exist?!) associated to agro-ecological data taken from remotely-sensed sources, such as Aviris images and LUCC.... but then again, if policy is to based at the meso-scale? How can we match downscaling with inconsistent national/district/ward level statistics if we are to construct tools to assist in facilitating effective coordination of, for instance, humanitarian aid?
Answers on a postcard please...
Yours faithfully in Geography,
Paul Georgie
--
BSc Hons Geography, MSc International Development & Management (NRM), MSc Geographical Information Science (2009/10)
Chris Nicholas
As Jawoo will attest with the DSSAT models, trying to model the effects of, say, intercropping on nitrogen uptake, carbon sequestration, or evapo-transpiration at the individual level is challenging enough; scaling it up to 'FutureClim' levels, along with WTO tariffs, a few random extreme weather events, game theory for carbon credit trading, etc is (or ought to be) the stuff of an NSF 'Grand Challenge' or equivalent.
At the global scale, there is the 'Community Climate System Model (CCSM)', as per: http://www.cesm.ucar.edu/models/ccsm4.0/ , with its 'community land model', that in turn has "Plant Functional Types", as per: http://www.cgd.ucar.edu/tss/clm/pfts/index.html . How these effectively map to other taxonomies, and adequately aggregate, is ... a challenge.
Certainly coupled economic-ecologic models, that are spatially explicit, is what this group is uniquely qualified to do. Interesting footnotes along the way are certainly various things done by CIFOR, such as 'FLORES', as per: http://www.simulistics.com/projects/flores/index.htm ; another is Bob Costanza's distributed 'Spatial Modeling Environment', as per: http://www.uvm.edu/giee/SME3/SModF.html , etc.
What language? (C#? Java? C++? Python?) What cloud computing environment? (Amazon? Azure? Hadoop? Globus?) Who knows...
Is there any 'gravitational' force to offset these 'centrifugal' forces? Do we wait for Google to save us with their 'Earth Engine'? (http://blog.google.org/2009/12/earth-engine-powered-by-google.html - a thin wrapper on IDRISI, plus DigitalGlobe and Landsat).
Not sure, but if there is a notion to do CSI-wide, coupled modelling, this is one forum to thrash it out. I'd certainly be willing to help author some proposals, and have various pieces sitting around from the ACP effort of last year.
Chris
From: Paul Georgie <paulg...@gmail.com>
To: cgia...@googlegroups.com
Cc: "Thornton, Philip (ILRI)" <P.THO...@cgiar.org>; p.j...@cigar.org; hei...@pik-potsdam.de
Sent: Thu, July 15, 2010 4:28:24 PM
Subject: Re: [CGIAR-CSI] FutureClim Update: 2080s added to the mix!
You received this message because you are subscribed to the Google Groups "CGIAR-CSI Community" group.
To post to this group, send email to cgia...@googlegroups.com.
To unsubscribe from this group, send email to cgiar-csi+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/cgiar-csi?hl=en.