Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Standard view
Full view
of .
Look up keyword
Like this
0 of .
Results for:
No results containing your search query
P. 1


Ratings: (0)|Views: 20,606|Likes:
Published by Phalippou Alexandre

More info:

Published by: Phalippou Alexandre on Dec 10, 2013
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less





 M anuscrip t p rep ared for J . H obbitlore  with version 5.0 of the L
  T  E X class cop erni- cus.cls.  D ate: 3 D ecember 20 13
  The C limate of Middle E arth
  R adagast the Brown
 1,2 1
  R hosgobel, nr. C arrock, M irkwood, M iddle E arth.
  T he C abot Institute, U niversity of B ristol, UK.
 Correspondence to:
 In this p ap er, I p resent  and discuss results from a climate model simulation of the `M iddle E arth' of elves, dwarves, and hobbits ( and not forgetting wizards such as my- self) . T hese are p ut into context by  also p resenting simulations of the climate of the `M odern E arth' of hu- mans, and of the `D inosaur E arth' , when  dinosaurs ruled the E arth 65 million years ago.Several asp ects of the M iddle E arth  simulation are discussed, including the imp ortance of p revailing wind drection for elvish sailing boats, the effect of heat and drought on the vegetation of M ordor, and the rain-shadow effects of the M istyM ountains.Ialsoidentifythose   p laces in the M odern E arth which have  the most similar climate to the re- gions of T he Shire and M ordor.  T he imp ortance of assessing `cli- mate sensitivity' ( the resp onse of the E arth to a doubling of at- mosp heric carbon dioxide concen- trations) is discussed, including the utility of modelling and recon- structing p ast climate change over timescales of millions of years. I also discuss the role of the Inter- governmental/Interkingdom P anel on  C limate C hange (IP  C  C   ) in assessing climate  change, and the resp onsibilities   p laced on p olicymakers.
 1 Introduction
  C omp uter models of the atmosp here,  land surface, and ocean are routinely used to p rovide forecasts of the weather and climate of the E arth.  T hey are based on our best theo- retical understanding of fluid mo- tion, p hysics, chemistry, and biol- ogy, written in the form of equa- tions, and then converted into a form which can be solved by a com-   p uter.  C limate models and models used to  make weather forecasts are very similar to each other, excep t that climate models typ ically simulate longer p eriods of time than weather
 2 R. Brown: C limate of Middle E arth
 models ( years to centuries as op   p osed  to days to weeks) , and therefore, due to limits on comp uter time and p ower, make p redictions at a lower sp atial resolution ( typ ical scale of hundreds of kilometers as op   p osed to kilome- ters) . C limate model p redictions are  an integral p art of p olitical and soci- etal p lanning for the coming decades to centuries, and the recent rep ort from the Intergovernmental P anel on  C limate C hange (IP  C  C   ) summarises many of  these future p redictions (
   ) . B ecause climate models are based on fundamental scientific understand- ing, they can be ap   p lied to many situ- ations. T hey are not designed solely for simulating the climate of the modern E arth, and, in theory, the same underlying science should ap   p ly to any time p eriod in the p ast. T he only  caveat is that in order to simu- late climates different from mod- ern, the user must p rovide some `boundary conditions' - map s or vari- ables which are not p redicted by the model. E xamp les include sp atial  map s of the height of the global terrain ( top ograp hy) and ocean dep th    ( bathymetry) , characteristics of rocks and soils, and concentrations of key atmosp heric constituents, such as ozone and carbon dioxide (  C  O
   ) . In addition, key p arameters such as the strength of the sun, and the radius and the rotation rate of the p lanet, also need to be p rovided to the model. Adap ting the model to simulate p ast time p eriods is p otentially very p ow-  erful because, in theory, we can know the `right' answer from ob- servations, and test the p erfor- mance of the models by comp aring their results with these obser- vations. F or time p eriods p rior to hu-  mans making careful observations of the weather, we rely on indi- rect observations of many asp ects of p ast climates, such as informa- tion from tree rings and ice cores, and fossils of p lants and animals. H ow- ever, p rovided that we understand the uncertainties and errors in these `p roxy' records of p ast climate change, and p rovided we also under- standandaccountforuncertainties in the boundary conditions we ap   p ly to  the model, we can make use of p ast p e- riods going back millions of years, to time p eriods when the E arth looked very different from the modern.In addition, by varying the to-   p ograp hy/ bathymetry, the rotation  rate and radius of the p lanet, and den- sity of the atmosp here, we can, in theory, use climate models to simu- late any p lanet, real or imagined.In this p ap er I p resent three climate  model simulations, of the M odern (   p re-  industrialised)E arth,oftheD inosaur  E arth ( a time p eriod called the L ate C re-  taceous, about 65 million years ago, just p rior to the extinction event which killed off the dinosaurs) , and M iddle E arth - the land of hobbits, elves, dwarves, wizards, and orcs.  T he aims of this p ap er are threefold:    ( 1) T o demonstrate the flexibility of climate models, arising from their basis in fundamental science.   ( 2) T o p resent the modelled climate of  M iddle E arth, and p rovide some light-  hearted discussion and interp reta- tions.   ( 3) T o discuss the strengths and lim- itations of climate models in gen-
  R. Brown: C limate of Middle E arth 3
 eral, by discussing ways in which the M iddle E arth simulations could be imp roved.
 2 Model descrip tion
I use a climate model develop ed at the UK M et O ffice, `H adC M 3L   ' , which is cap a-  ble of simulating the atmosp here, ocean, and land surface. In common with most climate models, H adC M 3L rep re-  sents the world in `gridbox' form, with a 3-dimensional network of boxes covering the surface and lay- ered to extend up to the top of the at- mosp here and down to the bottom of the ocean dep ths. T he size of eaach box is 3.75 degrees of longitude by 2.5 degrees of latitude, with a height dep endent on the distance from the  E arth' s surface - boxes situated near the surface of the E arth have a smaller height than those at the top of the atmosp here or bottom of the ocean. T his results in a `matrix' of boxes covering the world, with 96 boxes in the West-E ast direction, 73 boxes in N orth-South direction, 20 boxes deep in the ocean, and 19 high in the atmosp here ( a total of more than a quarter of a million boxes, although not all are used as some are effectively below the sea floor) .In this matrix, the fundamental equations of fluid motion in the at- mosp here and ocean are formulated and solved, with the additional comp lica- tionthattheE arthissp inningonits axis. E nergy is added to the system due to absorp tion of light and heat radiation emitted by the Sun, and energy leaves the system through emission of heat or reflection of light radiation into sp ace. All vari- ables in the model can be considered as average values over the volume of each gridbox, and so the climate model can not p rovide any information at a sp atial scale smaller than one gridbox ( so, for examp le, although it makes sense to talk about the modelled climate of the UK, or M ordor, the model can not give information about B ristol, or B ree) . H owever, in  reality there are many p rocesses which occur at a finer sp atial scale than that of a single gridbox. As such, models include `p arameterisa- tions' of sub-gridscale p rocesses, such as cloud formation, and small- scale atmosp heric turbulence, or ed- dies in the ocean. It is the rep resen- tation of these sub-gridscale p ro- cesses which brings uncertainty into climate modelling ( the equa- tions of fluid motion and thermody- namics themselves have been known and understood for several cen- turies) . As well as the atmosp here and ocean, the model includes a rep - resentation of the land and ocean surface, including p rocesses asso- ciated with sea-ice, soil moisture, and, in our p articular version of the model, the growth and distribution of vegetation.  T he model is given an initial state of all the variables which it p re- dicts ( for examp le temp erature, p res-  sure, wind sp eed, snow cover, ocean density) , and then the model is `run' forward in time, in step s of typ i- cally 10 to 30 minutes. Weather sys- tems develop and evolve, rain falls, the seasons come and go, and years of `model-time' p ass ( for my model, one year of model-time typ ically takes

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->