Special Topics in Atmospheric and Oceanic Sciences
Special Topics in Atmospheric and Oceanic Sciences ATOC 7500
Popular in Course
Popular in Marine Science
This 9 page Class Notes was uploaded by Jon Johns on Thursday October 29, 2015. The Class Notes belongs to ATOC 7500 at University of Colorado at Boulder taught by David Noone in Fall. Since its upload, it has received 8 views. For similar materials see /class/232049/atoc-7500-university-of-colorado-at-boulder in Marine Science at University of Colorado at Boulder.
Reviews for Special Topics in Atmospheric and Oceanic Sciences
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/29/15
NetCDF mapping and analysis of spacetime vanance How can we de ne climate Recall week 3 Lorenz attractor and week 10 QG model Data decomposition Consider spectral model QXyt qutYmXY Ym are spherical harmonics in longitude and mu More generally some variable Z Zxt Z atEx We are free to choose the basis functions E and derive the accompanying amplitudes a We will ndE are EOFS and a are the principal components Consider correlation Wish to choose basis that maximizes variance along primary axis Can show that this is just the eigenvectors of the correlationcovariance matrix EC XC Where C ZTZ E are the eigenvectors EOFs and 7 the eigenvalues Notice X2 is proportional to variance Recipe for PCA Eigenvector method EC AC Get some time space data Remove time mean Compute correlation scatter matrix Solve eigenvalue problem SVD method Z USVT Equivalent Get some time space data Remove time mean Perform SVD Notice S are singular values like eigenvalues and a property of SVG is that V is made of columns of eigenvectors of ZTZ and U the eigenvectors of 22quot Lab assignment Perform principal component analysis of monthly mean data from 19811990 Choose some region of the earth e g North America polar regions tropics Make plots of 1 Percentage variance explained by top few say 10 PCs 2 Amplitude versus time of first 3 PCs 3 Maps offirst 2 or 3 EOFs The data We will use data from the NCEP reanalysis This is monthly mean data from 19811990 thus a total of 120 months See NetCDF le dcnATOC7500projectdatancepmonthlync You do not need to copy this data read it from this directory A unix program ncdump shows contents of a NetCDF file tip give 7h option to see only header ncdump 7h ncepmonthlync netcdf ncepmonthly dimensions lon 6439 Dimensions lat 32 lev 3 time UNLIMITED 120 currently variables Z u n MORE VARIABLES HERE float lonlon lonlongname quotLongitudequot lonunits quotdegreeseastquot float latlat latlongname quotLatitudequot latunits quotdegreesnorthquot double timetime timelongname quotTimequot timeunits quotmonths since 01011981 000000quot float Ztime lev lat lon Zlongname quotGeopotential heightquot ts quotmquot Coordinate data Variables How to read data in IDL NetCDF les are binary and use ID codes to locate the data Open the file and return file ID ncid nodfopenfilename Get the ID of the variable varid ncdfvaridncid Z Read the data ncdfgetvarnoid varid 29 Close file ncdfclose ncid Hint you need to know the variable name Z first Get this from ncdump httpidlastrogsfc nasagovidlihtmlihelpNetCDFroutinesl0html Making map projection plots IDL does the work of figuring out the projections mapisetlat0lon0projectionitype Eg mapset30300orthographicisotropichorizon contour something mapcontinents mapiset has many options to play with Notice this is troublesome with lpmulti httpidlastrogsfc nasagovidlihtmlihelpproj ectionthtml Lab assignment Perform principal component analysis of monthly mean data from 19811990 Choose some region of the earth eg North America polar regions tropics Make plots of 1 Percentage variance explained by top few say 10 PCs 2 Amplitude versus time of rst 3 PCs 3 Maps of rst 3 EOFs Output as images and upload to Wiki Modi l PCA code to area weight the data to avoid area bias i e cosine of latitude Compare results Note 2d space could be longlat latlev or lon lev Key taskslessons for today lnterrogate NetCDF files ncdump Read NetCDF files in IDL Subset the data in IDL Perform PCA Contour plots with map projections See NetCDF file dcnATOC7500proj chdatancepmonthlync PCA routine and stu pcanalpro pcastubpro in dcnATOC7500week12 EOF of tropical sea surface temperature First EOF of 500 hPa height Annual modes The Nonhem IIemispheze annulaz mode The Seumem IIemisphexe annular mode Nnnhem The mean monthly pal component time series component mne sexies Thompson and Wallace 2001 Science 293 8589 Others 850 hPa streamfunction EOF 1 PNA EOF 2 NAO Paci c North American pattern North Atlantic Oscillation Are they real or statistical artifacts Summary Notice the NCEP data le has been interpolate to 64x32 grid and 3 layers of our climate model This may be useful for validation See IDL plotting script dcnATOC7500project