On predicting climate under climate change

Video thumbnail (Frame 0) Video thumbnail (Frame 187) Video thumbnail (Frame 1236) Video thumbnail (Frame 2894) Video thumbnail (Frame 3953) Video thumbnail (Frame 4649) Video thumbnail (Frame 5912)
Video in TIB AV-Portal: On predicting climate under climate change

Formal Metadata

Title
On predicting climate under climate change
Title of Series
Author
License
CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
2013
Language
English

Content Metadata

Subject Area
Abstract
Can today's global climate model ensembles characterize the 21st century climate in their own 'model-worlds'? This question is at the heart of how we design and interpret climate model experiments for both science and policy support. Using a low-dimensional nonlinear system that exhibits behaviour similar to that of the atmosphere and ocean, we explore the implications of ensemble size and two methods of constructing climatic distributions, for the quantification of a model's climate. Small ensembles are shown to be misleading in non-stationary conditions analogous to externally forced climate change, and sometimes also in stationary conditions which reflect the case of an unforced climate. These results show that ensembles of several hundred members may be required to characterize a model's climate and inform robust statements about the relative roles of different sources of climate prediction uncertainty.

Related Material

Video is accompanying material for the following resource
Phase (matter) Video Climate
Paper Hot working Electric power distribution Buick Century Buick Century Scale (map) Zugmaschine Year Global warming Climate Mobile phone Climate Railroad car Relative articulation Multiplizität Oceanic climate Adapter pattern Plant (control theory) Orbital period Musical ensemble Model building Angeregter Zustand Quantum Model building
Trajectory Stellar atmosphere Geokorona Bending (metalworking) Single (music) Climate Cogeneration Fulling Oceanic climate Climate change Musical ensemble Mint-made errors Control panel (engineering) Noise figure Flight simulator Atmospheric circulation Model building Sea surface temperature
Impact event Electric power distribution Year Command-line interface Single (music) Climate change Nyquist stability criterion Roll forming Adapter pattern Series and parallel circuits Musical ensemble Mint-made errors Video Control panel (engineering) Rail transport operations Model building Sea surface temperature
my name is just down on research with the consistent system analysis group at University together with Michael David Rankin Research Institute for the School of Economics this paper titled predicting climate and the climate change the paper questions how we design an answer on more experiments we ain't right contradictions on time scales tens to hundreds of we conclude that even for mobile work representation of reality cannot approaches to constantly or inadequate providing robust probabilities results are important how we want to find the and also how we use them adaptation decisions we began a letter with the central
guiding questions candidates today's global climate model of sometimes characterize the 21st century in their own model world to answer this question we 1st examine some of the conceptual issues relevant to climate prediction in the past most important and fundamental of the issues is how we define climate we contrast to competing definitions 1 common approach takes a distribution of whether over a period of time typically 30 years to define plant authors the from the nonlinear dynamical systems perspective we might consider an attractive to represent Parliament here the the oceans states at any point in time can be described by a quantum the attractor while climate is the entire distribution of states across the tractor income only if we took the 1st definition require time data populate our contributions and if we don't 2nd definition they require on some experiments from multiple initial conditions if we assume the 2 approaches we use the same distribution and we are making what we refer to as the car assumption using a low dimensional nonlinear
systems characteristics of the coupled atmosphere ocean system how we explore the consequences of adopting these 2 different conceptualizations Our model is the Ramseys full model which represents some aspects of the atmospheric circulation coupled some box in this figure the top left hand shows the attractor of model the ocean temperature and salinity variables in the top right panel the atmospheric X Y and Z under certain exporting conditions the 2 kind of the bottom here the attractive for predictions Faltings and in his later in our analysis small red spheres show the locations of the initial conditions are used to create the ensembles used master using single of
trajectories and initial condition ensembles we 1st examined the length of time it takes to reliably quantify the contributions of all other stationary forcing we find that ensembles are able to quantify the kind of model more quickly so far so good but in Britain climb on the climate change we are interested in the errors introduced by using time series data of a single small ensembles of simulations to quantify evolving climate this figure shows how the plant of
all ocean temperature and salinity variables involves under different forcing scenario so if you have all shown in blue how very large initial condition ensemble spreads out converges to a stable distribution line in orange vitally example single here we show that using data from single time series or small ensembles can lead errors of more than 20 percent when compared with the large ensemble data in the middle and bottom panels show scenarios parallel climate change where model parameter slowly changed over 100 years and then remains constant here we find that the errors resulting from it's important that I can be even larger furthermore even larger ensembles are available according to write distributions can converge the wrong 1 distribution luckily
considering ways to replicate it there in operation job cost considerably more expensive to however the results in show that users and how we design an inseparable experiments and have substantial impact on the probabilities we divide the by this is crucial information form policies and guide adaptations the the status of the options that they were going to be like any additional information based on unhesitating on thank you very much for this thing I think you really think
Feedback