EGU 2012-7570 CL5.1/GM2.5 TOYE LEMIEUX BLAYO Objectives Mathematical formalism Data flow chart Background inputs Background error covariance matrix Observation inputs Outputs Input and output files extracts On going work DatIce code 1/12 EGU 2012 Toward unified ice core chronologies with the DatIce tool Habib TOYE MAHAMADOU KELE B´ en´ edicte LEMIEUX-DUDON, ´ Eric BLAYO Jean Kuntzmann Laboratory INRIA and University of Grenoble joint work with LGGE laboratory April ,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
EGU 2012-7570CL5.1/GM2.5
TOYELEMIEUX
BLAYO
Objectives
Mathematicalformalism
Data flow chart
Backgroundinputs
Background errorcovariance matrix
Observationinputs
Outputs
Input and outputfiles extracts
On going work
DatIce code
1/12
EGU 2012Toward unified ice core chronologies with the
DatIce tool
Habib TOYE MAHAMADOU KELEBenedicte LEMIEUX-DUDON, Eric BLAYO
Jean Kuntzmann LaboratoryINRIA and University of Grenoblejoint work with LGGE laboratory
April ,
EGU 2012-7570CL5.1/GM2.5
TOYELEMIEUX
BLAYO
Objectives
Mathematicalformalism
Data flow chart
Backgroundinputs
Background errorcovariance matrix
Observationinputs
Outputs
Input and outputfiles extracts
On going work
DatIce code
2/12Objectives
I Present the DatIce software toolI This tool implements an inverse method for multiple ice
cores datingI It provides an improved chronology, using a prior
chronology and independent observationsI This tool is available to the scientific community
I This talk is NOT a presentation on paleoclimatologyDatIce
EGU 2012-7570CL5.1/GM2.5
TOYELEMIEUX
BLAYO
Objectives
Mathematicalformalism
Data flow chart
Backgroundinputs
Background errorcovariance matrix
Observationinputs
Outputs
Input and outputfiles extracts
On going work
DatIce code
3/12Mathematical formalism
We want to use data from previous chronology andobservation markers to get a new chronology.
EGU 2012-7570CL5.1/GM2.5
TOYELEMIEUX
BLAYO
Objectives
Mathematicalformalism
Data flow chart
Backgroundinputs
Background errorcovariance matrix
Observationinputs
Outputs
Input and outputfiles extracts
On going work
DatIce code
4/12Mathematical formalism
X : unknwon vector representing a chronology
J(X ) =˛˛
X − Xb˛˛2
B| {z }misfit to the background
+ ||h(X )− Yo||2R| {z }misfit to observations
We are looking for X a, ”the best new chronology” that minimizes the J function.