A new algorithm for the downscaling of 3

  • Slides: 20
Download presentation
A new algorithm for the downscaling of 3 -dimensional cloud fields Victor Venema Sebastián

A new algorithm for the downscaling of 3 -dimensional cloud fields Victor Venema Sebastián Gimeno García Clemens Simmer

Applications § Downscaling 3 D CRM/NWP model fields § Downscaling of 2 D satellite

Applications § Downscaling 3 D CRM/NWP model fields § Downscaling of 2 D satellite measurements § Coarse mean LWC § Coarse cloud fraction Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Requirements downscaling method § Nonlinear processes – Sub (coarse) scale distribution – IPA-bias: if

Requirements downscaling method § Nonlinear processes – Sub (coarse) scale distribution – IPA-bias: if you average instead of (ir)radiances § Non-local processes – For example spatial correlations – 3 D bias: ignore horizontal photon transport to low Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Downscaling - Cumulus Original Coarse means § High resolution original => – Coarse means

Downscaling - Cumulus Original Coarse means § High resolution original => – Coarse means – No clear subpixels § 2 coarse fields No. clear subpixels – Input downscaling § Real application start with coarse fields § Compare highresolution fields – Physical – Radiative Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Downscaling - Cumulus Coarse means § High resolution original => – Coarse means –

Downscaling - Cumulus Coarse means § High resolution original => – Coarse means – No clear subpixels § 2 coarse fields No. clear subpixels Surrogate – Input downscaling § Real application start with coarse fields § Compare highresolution fields – Physical – Radiative Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Downscaling - Cumulus Original Coarse means § High resolution original => – Coarse means

Downscaling - Cumulus Original Coarse means § High resolution original => – Coarse means – No clear subpixels § 2 coarse fields No. clear subpixels Surrogate – Input downscaling § Real application start with coarse fields § Compare highresolution fields – Physical – Radiative Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Cumulus validation data § § Diurnal cycle of Cu Land (ARM) 51 fields High

Cumulus validation data § § Diurnal cycle of Cu Land (ARM) 51 fields High resolution – 64 x 64 pixels – Horizontal resolution 100 m § Coarse resolution – 16 x 16 – Horizontal resolution 400 m § Nc = 300 cm-3 Brown, A. R. , R. T. Cederwall, A. Chlond, P. G. Duynkerke, J. C. Golaz, M. Khairoutdinov, D. C. Lewellen, A. P. Lock, M. K. Mac. Vean, C. H. Moeng, R. A. J. Neggers, A. P. Siebesma and B. Stevens, 2002. Large-eddy simulation of the diurnal cycle of shallow cumulus convection over land, Q. J. R. Meteorol. Soc. , 128(582), 1075 -1093.

Stratocumulus validation data § § Dissolving broken Sc Ocean (ASTEX) 29 fields High resolution

Stratocumulus validation data § § Dissolving broken Sc Ocean (ASTEX) 29 fields High resolution – 200 x 200 pixels – Horizontal resolution 50 m § Coarse resolution – 20 x 20 – Horizontal resolution 500 m § Nc = 200 cm-3 Chosson, F. , J. -L. Brenguier and L. Schüller, "Entrainment-mixing and radiative Transfer Simulation in Boundary-Layer Clouds", J Atmos. Res.

Algorithm § Preparations – Calculate power spectrum coarse LWC field – Extrapolate spectrum to

Algorithm § Preparations – Calculate power spectrum coarse LWC field – Extrapolate spectrum to smaller scales § Main iterative loop – Adjust to the extrapolated spectrum – Adjust to the coarse fields – Remove jumps at edges of coarse field

Algorithm – flow diagram Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Algorithm – flow diagram Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Extrapolation power spectrum § Algorithm works with any power spectrum § Cumulus clouds –

Extrapolation power spectrum § Algorithm works with any power spectrum § Cumulus clouds – Assumption: § Intermediate to small scales are fractal § follow power law (Variance=akb) – Linear regression in log-log spectrum – Fitting range: § small scales of coarse field (intermediate scales full field) § Stratocumulus cloud – Not fractal at intermediate scales – Assumption: § Shape power spectrum same for all clouds – Computed an average isotropic spectrum over all clouds – Scaled by average variance at intermediate scales Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Example 3 D fields Original Extrapolated Surrogate Coarse field Cumulus Stratocumulus Victor. Venema@uni-bonn. de,

Example 3 D fields Original Extrapolated Surrogate Coarse field Cumulus Stratocumulus Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Example 3 D fields Original Extrapolated Surrogate Coarse field Cumulus Stratocumulus Victor. Venema@uni-bonn. de,

Example 3 D fields Original Extrapolated Surrogate Coarse field Cumulus Stratocumulus Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Scatterplot irradiances Cu Reflectance SZA 0° Reflectance SZA 60° Transmittance SZA 60° Two originals

Scatterplot irradiances Cu Reflectance SZA 0° Reflectance SZA 60° Transmittance SZA 60° Two originals Extrapolated surrogate Coarse field Interpolated field Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Scatterplot irradiances Sc Reflectance SZA 0° Reflectance SZA 60° Transmittance SZA 60° Two originals

Scatterplot irradiances Sc Reflectance SZA 0° Reflectance SZA 60° Transmittance SZA 60° Two originals Extrapolated surrogate Coarse field Interpolated field Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

RMS relative difference Rel. Diff. = (Field-Orig)/Orig Cumulus Field Stratocumulus Reflectance Transmittance Second original

RMS relative difference Rel. Diff. = (Field-Orig)/Orig Cumulus Field Stratocumulus Reflectance Transmittance Second original 0. 01 0. 002 0. 0002 Coarse field 0. 52 0. 0271 0. 144 0. 0115 Interpol. field 0. 99 0. 0540 0. 208 0. 0157 Extrapolated spect. 0. 07 0. 0032 0. 038 0. 0032 Fractal spectrum 0. 07 0. 0042 0. 020 0. 0009 Exact spectrum 0. 01 0. 0002 0. 007 0. 0005 Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

RMS relative difference Rel. Diff. = (Field-Orig)/Orig Cumulus Field Stratocumulus Reflectance Transmittance Second original

RMS relative difference Rel. Diff. = (Field-Orig)/Orig Cumulus Field Stratocumulus Reflectance Transmittance Second original 0. 01 0. 002 0. 0002 Coarse field 0. 52 0. 0271 0. 144 0. 0115 Interpol. field 0. 99 0. 0540 0. 208 0. 0157 Extrapolated spect. 0. 07 0. 0032 0. 038 0. 0032 Fractal spectrum 0. 07 0. 0042 0. 020 0. 0009 Exact spectrum 0. 01 0. 0002 0. 007 0. 0005 Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Conclusions § Downscaling algorithm works – Large improvement for irradiances compared to coarse cloud

Conclusions § Downscaling algorithm works – Large improvement for irradiances compared to coarse cloud fields § Extrapolation is a significant error source – Low number of pixels in coarse fields – Best extrapolation method is application dependent Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Outlook § § Importance of the coarse cloud fraction field Include a distribution for

Outlook § § Importance of the coarse cloud fraction field Include a distribution for the anomalies Wavelets, increment distributions? Applications – Downscaling CRM/NWP model fields § Anomalies, small-scale spectrum from LES or observations – Downscaling of satellite measurements § Coarse LWP fields § High resolution in situ LWC, Reff measurements Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Outlook § § Importance of the coarse cloud fraction field Include a distribution for

Outlook § § Importance of the coarse cloud fraction field Include a distribution for the anomalies Wavelets, increment distributions? Applications – Downscaling CRM/NWP model fields § Anomalies, small-scale spectrum from LES or observations – Downscaling of satellite measurements § Coarse LWP fields § High resolution in situ LWC, Reff measurements Thank you for your attention! Victor. Venema@uni-bonn. de, http: //www. meteo. uni-bonn. de/venema

Costeffective dynamical downscaling An illustration of downscaling CESMCosteffective dynamical downscaling An illustration of downscaling CESM
Downscaling Climate Variables Downscaling Inferring climate variations onDownscaling Climate Variables Downscaling Inferring climate variations on
Domino algorithm Domino algorithm Domino algorithm Domino algorithmDomino algorithm Domino algorithm Domino algorithm Domino algorithm
Algorithm Algorithm Defination An Algorithm is a finiteAlgorithm Algorithm Defination An Algorithm is a finite
Recursive Algorithm Recursive Algorithm Recursive Algorithm R LeongRecursive Algorithm Recursive Algorithm Recursive Algorithm R Leong
Algorithm Cost Algorithm Complexity Algorithm Cost LB BackAlgorithm Cost Algorithm Complexity Algorithm Cost LB Back
Algorithm Cost Algorithm Complexity Algorithm Cost LB BackAlgorithm Cost Algorithm Complexity Algorithm Cost LB Back
Context Dijkstra algorithm BellmanFord algorithm A algorithm FloydWarshallContext Dijkstra algorithm BellmanFord algorithm A algorithm FloydWarshall
Dijkstras Algorithm Dijkstras Algorithm This algorithm finds theDijkstras Algorithm Dijkstras Algorithm This algorithm finds the
Algorithm Cost Algorithm Complexity Algorithm Cost LB BackAlgorithm Cost Algorithm Complexity Algorithm Cost LB Back
Shortest Path Algorithm Dijkstras Algorithm BellmanFord Algorithm 8Shortest Path Algorithm Dijkstras Algorithm BellmanFord Algorithm 8
Shortest Path Algorithm Dijkstras Algorithm BellmanFord Algorithm 8Shortest Path Algorithm Dijkstras Algorithm BellmanFord Algorithm 8
Algorithm Analysis and Recursion 1 Algorithm An algorithmAlgorithm Analysis and Recursion 1 Algorithm An algorithm
Algorithm Analysis and Recursion 1 Algorithm An algorithmAlgorithm Analysis and Recursion 1 Algorithm An algorithm
Tuned Boyer Moore Algorithm Raita Algorithm Horspool AlgorithmTuned Boyer Moore Algorithm Raita Algorithm Horspool Algorithm
Shortest Path Algorithm Dijkstras Algorithm BellmanFord Algorithm 8Shortest Path Algorithm Dijkstras Algorithm BellmanFord Algorithm 8
Recursive Algorithm Recursive Algorithm Recursive Algorithm R LeongRecursive Algorithm Recursive Algorithm Recursive Algorithm R Leong
Deutschs Algorithm DeutschJozsa Algorithm Deutschs algorithm fx ANDDeutschs Algorithm DeutschJozsa Algorithm Deutschs algorithm fx AND
Algorithm Analysis 1 Algorithm An algorithm is aAlgorithm Analysis 1 Algorithm An algorithm is a