Variable to be converted BofA=C1 A false 1 false int readgrib2 2 2 false Grids over which median will be derived grids medianover2 3 var' medianover3 false median over the given grids medianover1 var false 1 Variable from which median will be derived nx 2 false number of longitude points ny number of latitude points false 3 false var' Variable after conversion 4 false 1 Variable to be converted var gaussianlat gaussianlat returns gaussian grid with n points. Returns gaussian grid with n points n gaussianlat1 1 false false Gaussian grid with n points gaussianlat2 grid 2 Applies zonal derivative to variable in spectral coordinates partialeast applies zonal derivative to variable in spectral coordinates Variable after zonal derivative in spectral coordinates has been applied var' 2 partialeast2 false false Variable to which zonal derivative will be applied 1 var partialeast partialeast changetruncation changetruncation 1 false changetruncation1 var Variable to be changed false var' changetruncation3 New spectral truncation level 3 false nsp 2 changetruncation2 Changes variable truncations to nsp changes variable truncations to nsp laplacian applies Laplacian to variable in spectral coordinates New variable to which Laplacian has been appllied false laplacian2 var' 2 Variable to which Laplacian will be applied 1 laplacian1 false var Applies Laplacian to variable in spectral coordinates laplacian number of latitude points false 3 ny tsptogau3 converts spectral coefficients to lon/lat tsptogau 2 number of longitude points nx tsptogau2 false var' Variable converted to lon/lat false 4 tsptogau4 tsptogau Converts spectral coefficients to lon/lat var false Variable to be converted tsptogau1 1 Spectral Transformation false Variable converted to spectral coefficients velocity var' 3 2 false spectral truncation nwave Variable to be converted var false 1 gautotspUV Converts lon/lat to spectral coefficients using spectral harmonics for velocity tsptogauUV false 2 nx number of longitude points Variable to be converted 1 var false 4 false var' Variable converted to lon/lat velocities ny number of latitude points false 3 Converts spectral coefficients to lon/lat using spectral harmonics for velocity Variable after meridional derivative applied partialnorth2 2 var' false false Variable to which meridional derivative will be applied to var partialnorth1 1 partialnorth applies meridional derivative to variable in spectral coordinates Applies meridional derivative to variable in spectral coordinates partialnorth Lon/Lat false gautotsp1 var 1 spectral truncation nwave false gautotsp2 2 3 var' gautotsp3 false Spectral coefficients Converts lon/lat to spectral coefficients converts lon/lat to spectral coefficients gautotsp gautotsp invlaplacian false 2 Inverse laplacian applied to stream in spectral coordinates invlaplacian2 var' false var Varibable to which inverse laplacian will be applied 1 invlaplacian1 invlaplacian applies inverse laplacian to variable in spectral coordinates Applies inverse laplacian to variable in spectral coordinates tsptogauR Converts spectral coefficients to lon/lat Combines two grids (i.e., independent variables) that were created by the splitstreamgrid function (i.e., undoes splitstreamgrid) 3 same as <i>var</i> except now dependent on <i>grid_combined</i> instead of <i>grid</i> and <i>grid</i>2 false varout Combines two grids (i.e., independent variables) that were created by the splitstreamgrid function (i.e., undoes splitstreamgrid) SOURCES .NOAA .NCEP .EMC .CMB .GLOBAL .Reyn_SmithOIv2 .monthly .sst <br>T (Jan 1982) (Dec 2003) RANGE <br>T 12 splitstreamgrid <br>T unsplitstreamgrid 4 false grid containing combined values of <i>grid</i> and <i>grid</i>2 grid_combined var 1 false variable dependent on grids to be combined independent variable to be sampled along grid 3 false false 4 sampled variable -- all values of grid found in var2 have been dropped. var1-sampled samples variable by keeping only what is missing from a second variable false variable to be sampled 1 var1 3 /SAMPLE_MISSING publicproc: dup type /arraytype eq { dup 0 get totype /gridtype eq } if { 0 get .name 2 index 1 index getgridbyname nip 3 1 roll getgridbyname nip } if {mydataset initial_grid remove_grid}inputs mydataset initial_grid /districtmap unitmatrix streamgrids pop remove_grid SAMPLE [streamgrids pop]sum 0 maskgt streamgrids initial_grid .name cvlit renameGRID SELECT 1output :publicproc var2 2 false where list of values to remove is taken from Regrids a variable by averaging. Indicated grids of one variable are regridded to match those of another variable. The function is commonly used when comparing multiple variables that are dependent on different spatial grids (see correlation example below). SOURCES .NOAA .NCEP .CPC .Merged_Analysis .monthly .v0407 .ver2 .prcp_est <br>SOURCES .CDIAC .tr051 .precipitation .anomalies <br>[X Y] regridAverage <p> OR <p> SOURCES .NOAA .NCEP .CPC .Merged_Analysis .monthly .v0407 .ver2 .prcp_est <br>T (Jan 1980) (Dec 2003) RANGE <br>SOURCES .NOAA .NCDC .ERSST .version2 .SST <br>T (Jan 1980) (Dec 2003) RANGE <br>[X Y]0.5 regridAverage <br>[T]correlate <p>Note that in the second example, the <i>minfrac</i> value of 0.5 requires that at least half of the input data that go into a particular output gridpoint have non-missing values in order for that gridpoint to receive a non-missing value in the regridding (i.e., keeps land from expanding outward). Regridding regridLB2 false Variable to be regridded 2 var2 false Regridded variable regridLB5 5 var2' Regrids variable2 to match variable1 by using left bounds, i.e. grid is ordered, variable2's grid is a subset of variable1's grid, the values of the variable1 grid are bounded by the values of the variable2 grid, and the variable2 values that are on the left of that pair are used as the matching values regridLB4 4 false Unchanged variable1 var1 Variable to be compared to 1 regridLB1 false var1 ivar 3 regridLB3 Independent variable to be applied to to variables false regridLB regrids variable2 to match variable1 by using left bounds, i.e. grid is ordered, variable2's grid is a subset of variable1's grid, the values of the variable1 grid are bounded by the values of the variable2 grid, and the variable2 values that are on the left of that pair are used as the matching values. regridLB <br> <b><u>Description</u></b><br><br><b>GRID</b> regrids an existing independent variable (grid) associated with a data set or variable with a newly-defined evenly-spaced grid based upon user-defined limits and spacing.<br><br>In the function arguments &quot;var&quot; is a data variable that depends on the independent variable &quot;grid&quot;, &quot;lowbound&quot; is the minimum value of the new grid being defined, &quot;step&quot; is the evenly-spaced interval of the new grid in the units of &quot;grid&quot;, and &quot;highbound&quot; is the upper limit of the new grid.<br><br><b><u>Example</u></b><br><br>In the following example, a gridded data set of monthly mean short wave radiances is spatially regridded globally (from -180&#176; to 179&#176; in longitude and -90&#176; to 90&#176; in latitude) from 2.5&#176; lat/lon grid spacing to 1&#176; lat/lon spacing. Bi-linear interpolation is used to regrid from the coarser to the finer grid.<br><br><font color="#008000">SOURCES .ERBE .global .monthly .srb<br>X -180 1 179 GRID<br>Y -90 1 90 GRID</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.ERBE/.global/.monthly/.srb/X/-180/1/179/GRID/Y/-90/1/90/GRID/">Live Example Link</a> regridvar <i>var</i> regridded onto newly-defined grid <p> Data at new grid points are based on linear interpolation of original <i>var</i>. 6 false false grid grid (i.e., independent variable) on which new grid specifications (e.g., lowbound, step, highbound) are to be applied 2 1 var variable to be regridded false 3 false lowbound lower limit of new grid definition Regrids a variable onto a newly-defined, evenly-spaced grid (i.e., independent variable) 4 step width (in units of <i>grid</i>) between new grid points <p> If n is not an integer, where n=(<i>highbound</i>-<i>lowbound</i>)/<i>step</i>, then <i>step</i> is assigned closest value that allows n to be an integer. false 5 false highbound upper limit of new grid definition GRID Regrids a variable onto a newly-defined, evenly-spaced grid (i.e., independent variable) var2' regridLinear7 Variable after regridding 7 false 4 Wght false regridLinear4 regridLinear2 Variable to be regridded false 2 var2 Wmin true 5 regridLinear5 3 ivar' Independent variable to be applied to variables false regridLinear3 regridLinear Regrids by averaging regridLinear Unchanged variable regridLinear6 6 var1 false false 1 regridLinear1 var1 Variable to be regridded regrids by two point linear interpretation 3 ship track false transit3 Y(S) false transit5 Returns the position on the given grid of the ship track 5 D(S) given a variable D(X,Y) and a ship track X(S), Y(S), returns D(S). X(S) ship track transit2 false 2 Given a variable D(X,Y) and a ship track X(S), Y(S), returns D(S). transit transit false 1 D(X,Y) transit1 4 transit4 false S minfrac 5 true Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the average is calculated regardless of the amount of data present in domain. variable whose grid(s) are to be regridded 2 false var2 Regrids a variable by averaging. Indicated grids of one variable are regridded to match those of another variable. The function is commonly used when comparing multiple variables that are dependent on different spatial grids (see correlation example below). weights true weights to be used in the averaging. Note: not generally needed unless regridding to a very coarse spatial grid. 4 false var2regridded Same as <i>var2</i> except it is now dependent on <i>grids</i> as they are defined in <i>var1</i>. 6 3 grid(s) of <i>var2</i> to be regridded to match those of <i>var1</i>. Note: use caution when applying the function to temporal grids to make sure that the desired output is actually returned, particularly when regridding a coarse temporal grid to a relatively fine temporal grid. grids false regridAverage 1 variable (i.e, its grids) on which regridding is based false var1 false dataset 1 dataset to be altered true newname character argument string name to use for independent variable 3 false var 1D dependent variable to be used as the new independent variable 2 new dataset with var as independent variable 4 false dataset' use_as_grid uses a 1D variable to replace the independent variable on which it depends <p style="margin-top: 0"> dup type /stringtype eq {cvn} if </p> <p style="margin-top: 0"> dup type /nametype eq {3}{2} ifelse </p> <p style="margin-top: 0"> /use_as_grid publicproc: </p> <p style="margin-top: 0"> dup type /nametype eq { cvx exch 1 object exch /name exch def } if </p> <p style="margin-top: 0"> streamgrids exch </p> <p style="margin-top: 0"> datatype /stringtype eq {toname}if </p> <p style="margin-top: 0"> ndim RECHUNK </p> <p style="margin-top: 0"> name units ordered 4 -1 roll getrealization NewGRID </p> <p style="margin-top: 0"> 3 -1 roll 5 object 1 index name exch def 3 1 roll </p> <p style="margin-top: 0"> replaceGRID </p> <p style="margin-top: 0"> :publicproc </p> SAMPLE_MISSING unsplitstreamgrid 2 grid false grid that was split with splitstreamgrid (see <i>grid_in</i> and <i>gridout</i> in splitstreamgrid function documentation) high High value on ivar false 5 false 7 New ivar evenly spaced from low to high with step var' ivar false 2 Independent variable to be applied evengridAverage step Step between low and high on new ivar false 4 var 1 Variable to be regridded false Low value on ivar 3 false low Regrids a variable to a new grid by averaging true 6 Wmin false laggedcovariance4 lags 4 3 laggedcovariance3 false period 6 laggedcovariance6 cov false lagged covariance 5 false laggedcovariance5 Egrid Computes lagged covariance $C(E,E,S,L)$. If Tgrid is a variable, uses the first grid as the time grid laggedcovariance laggedcovariance false 2 laggedcovariance2 Tgrid computes lagged covariance $C(E,E,S,L)$. If Tgrid is a variable, uses the first grid as the time grid. laggedcovariance1 false 1 var magnitude of vector with components of <i>var1</i> and <i>var2</i>. <p><i>vectormag</i> is dependent on all of the grids on which <i>var1</i> and <i>var2</i> are dependent. vectormag 3 false SOURCES .NOAA .NCEP-NCAR .CDAS-1 .MONTHLY .Intrinsic .PressureLevel .u <br>SOURCES .NOAA .NCEP-NCAR .CDAS-1 .MONTHLY .Intrinsic .PressureLevel .v <br>P 850 VALUE <br>mag 1 var1 first component of vector false Calculates the vector magnitude 2 second component of variable var2 false mag Calculates the vector magnitude dens Pressure P false dens3 3 unecso '81 density (insitu) of seawater oceanic equation of state false Temperature 1 T dens1 <p style="margin-top: 0"> <br> <b><u>Description</u></b><br><br><b>dens</b> calculates the density of seawater (in g/cm³) given input variables temperature (T), salinity (S), and pressure (P), in that order, using the international equation of state for seawater, IES80 (UNESCO 1981).<br><br><b><u>Reference</u></b><br><br>UNESCO, 1981: Tenth report of the joint panel on oceanographic tables and standards. UNESCO Tech. Paper in Marine Science 36, 25pp.<br><br><b><u>Example</u></b><br><br>The following example calculates the density of seawater based on samples of measured temperature, salinity, and pressure from the full data set of GEOSECS observational profiles.<br><br><font color="#008000">SOURCES .GEOSECS .TEMP </font> </p> <p style="margin-top: 0"> <font color="#008000">SOURCES .GEOSECS .SAL </font> </p> <p style="margin-top: 0"> <font color="#008000">SOURCES .GEOSECS .PRESS </font> </p> <p style="margin-top: 0"> <font color="#008000">dens</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.GEOSECS/.TEMP/SOURCES/.GEOSECS/.SAL/SOURCES/.GEOSECS/.PRESS/dens/">Live Example Link</a> </p> S Salinity false dens2 2 dens false dens4 Unecso '81 density (insitu) of seawater density 4 Unecso '81 density (insitu) of seawater :cressman cressman: start of cressman scheme parameters 5 Performs a cressman objective analysis to create gridded data from station data 2 latitude latitude coordinates for stations used in the analysis griddeddata 6 returned gridded data data values at the stations data 3 IWMO 4 station ids longitude coordinates for stations used in the analysis longitude 1 <br> <u><b>Detailed Syntax</b></u><br><br><i>longitudes</i><br><i>latitudes</i><br><i>data</i><br><b>[</b><i>IWMO</i><b>]</b> <b>cressman:</b><br><i>n</i> <b>minstns</b><br><b>X</b> <i>xmin</i> <i>xmax</i> <i>xstep</i> <b>RANGEEDGESTEP</b><br><b>Y</b> <i>ymin</i> <i>ymax</i> <i>ystep</i> <b>RANGEEDGESTEP</b><br><i>n1</i> <b>pass1</b><br><i>n2</i> <b>pass2</b><br><i>n3</i> <b>pass3</b><br><b>:cressman</b><br><br><u><b>Input</b></u><br><br><i>longitudes</i> is a stream of longitude coordinates for stations used in the analysis<br><i>latitudes</i> is a stream of latitude coordinates for stations used in the analysis<br><i>data</i> is a stream of data values at the stations<br><i>IWMO</i> is the name of the grid of station identifiers<br><br><u><b>Parameters</b></u><br><br><i>n </i><b>minstns</b> : <i>n</i> is the number of station data points that must be included within the influence radius of a gridpoint for an analysis value to be calculated for that gridpoint. The default value is 3.<br><br><b>X</b> <i>xmin</i> <i>xmax</i> <i>xstep</i> <b>RANGEEDGESTEP</b> : specification of the east-west domain of the analysis to be produced, in terms of minimum and maximum longitude values and the grid step, in degrees. The default values for <i>xmin</i>, <i>xmax</i>, and <i>xstep</i> are -180., 180., and 2., respectively.<br><br><b>Y</b> <i>ymin</i> <i>ymax</i> <i>ystep</i> <b>RANGEEDGESTEP</b> : specification of the north-south domain of the analysis to be produced, in terms of minimum and maximum latitude values and the grid step, in degrees. The default values for <i>ymin</i>, <i>ymax</i>, and <i>ystep</i> are -90., 90., and 2., respectively.<br><br><i>n1</i> <b>pass1</b>, <i>n2</i> <b>pass2</b>, <i>n3</i> <b>pass3</b> : <i>n1</i>, <i>n2</i>, and <i>n3</i> are the factors by which the average minimum station separation distance (calculated within the routine) is multiplied on each pass of the cressman scheme (3 passes are automatically done), to establish the influence radius for each pass. The default values for the first, second, and third passes are 4, 2.5, and 1.5, respectively.<br><br><u><b>Description</b></u><b> </b><br><br><b>cressman</b> performs a Cressman (1959) objective analysis of input station data onto a user-defined latitude-longitude grid. A three-pass successive correction scheme is used here, allowing for the influence radius to be tightened on each pass. The influence radius (R) for each pass is calculated as the product of a user-defined constant and the average distance between each station and its nearest neighboring station. On each pass of the analysis, for each gridpoint in the analysis domain, an isotropic distance-based weight is calculated for each station within the influence radius of the gridpoint. This weight is calculated as follows: W = (R² - r²)/(R² + r²) where R = influence radius and r = station-to-gridpoint distance. Distances are calculated along a great circle path. The weighting function is illustrated in the following graph:<br><br><img alt="Image of cressman weighting function" src="http://iri.columbia.edu/~mbell/cress_weight.jpg"><br><br>The analysis value on the third and final pass at each gridpoint is calculated as:<br><br>Z₃ = Z₂ + sum[W*(zo_j - zb_j)]/sum[W]<br><br>where Z₂ is the value of the analysis on the previous pass, W is the distance-based weight for each station within the influence radius of the gridpoint, zo_j is the observed value at station j, zb_j is the background value at the station location j calculated via bilinear interpolation from the analysis on the previous pass, and the sums are calculated over the number of stations within the influence radius of each gridpoint.<br><br>If the minimum station number requirement specified in the <b>minstns</b> parameter is not met or exceeded, a missing value will be assigned to the gridpoint. This allows the user to restrict the region to which valid analysis values are assigned based upon a simple measure of station density.<br><br><u><b>References</b></u><br><br>Cressman, G. P., 1959: An operational objective analysis system. <i>Mon. Wea. Rev.</i>, <b>87</b>, 367-374.<br><br><u><b>Example</b></u><br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .CAMS .station .temperature </font><br><font color="#008000">lon lat temp [IWMO] cressman: </font><br><font color="#008000">3 minstns </font><br><font color="#008000">X -180. 180. 2. RANGEEDGESTEP </font><br><font color="#008000">Y -90. 90. 2. RANGEEDGESTEP </font><br><font color="#008000">4. pass1 </font><br><font color="#008000">2.5 pass2 </font><br><font color="#008000">1.5 pass3 </font><br><font color="#008000">:cressman</font><br><br><a href=" http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.CAMS/.station/.temperature/lon/lat/temp%5BIWMO%5Dcressman:/3/minstns/X/-180./180./2./RANGEEDGESTEP/Y/-90./90./2./RANGEEDGESTEP/4./pass1/2.5/pass2/1.5/pass3/:cressman/">Live Example Link</a><br> Performs a cressman objective analysis to create gridded data from station data variable containing missing values to be replaced <p> Note that missing values must be identified with the <i>missing_value</i> attribute. If they are not, then the attribute can be defined as in the example below. var 1 false <i>var</i> with all missing values identified by the <i>missing_value</i> attribute replaced by <i>value</i> varfull 3 false 2 false value value or variable that will replace missing values Replaces missing values in a variable with a selected value (or variable) Replaces missing values in a variable with a selected value (or variable) replaceNaN SOURCES .NOAA .NODC .WOA01 .Grid-1x1 .Monthly .an .salinity <br>/missing_value -99.9999 def <br>0 replaceNaN variable in categorical form classes false 3 complete disjunctive form classify Classifies data into categories, i.e. labels ranges of values. false 4 weights output. There is an additional grid consisting of the N+1 names, and the values are 0, 1, or missing depending on whether the data was between the values given in the <i>classify</i> number set. This variable is sometimes referred to as being in <i>complete disjunctive form</i>. <br> <b><u>Description</u></b><br><br><b>classify</b> is used to assign ranges of values from a variable into user-defined classes. Given a variable with a given range of values, the classify statement accepts a list of alternating class names and constants which define the boundaries between the classes within that range. As a result, a new grid composed of the defined classes is created, and the values from the input variable are transformed into flags of 0 (not a member of the class), 1 (is a member of the class), or NaN (not a number -- missing). This is best illustrated with an example.<br><br><b><u>Examples</u></b><br><br><font color="#008000">SOURCES .KAPLAN .Indices .NINO3 .avOS<br>T (Jan 1901) (Dec 1990) RANGE<br>T 3 boxAverage<br>[T]percentileover<br>{LaNina 0.2 Neutral 0.8 ElNino}classify</font><br><br>This example first takes non-overlapping 3-month seasonal averages of sea surface temperature anomalies (SSTA) from the NINO3 region of the equatorial Pacific Ocean over the period January 1901 to December 1990. This gives a single time series of seasonal sea surface temperature anomalies. The first time step is Jan-Mar 1901, the second is Apr-Jun 1901, and so on until Oct-Dec 1990.<br><br>Then, the SSTA values are converted into percentiles, from 0. to 1., using <b>[T]percentileover</b>. The most negative SSTA values in the distribution are assigned a value near zero, and the most positive values are assigned a value near one, with intermediate values ranging between these extremes.<br><br>The next line comprises the <b>classify</b> statement and its parameters. The class names and the boundaries between them are placed within the curly braces. Since the input variable is composed of percentiles that range between 0. and 1. the class boundaries should fall within this range. In this case, any values below 0.2 (in the lowest 20% of the distribution) are classified as 'LaNina', values between 0.2 and 0.8 are classified as 'Neutral', and values from 0.8 upward (in the upper 20% of the distribution) are classified as 'ElNino'.<br><br>Whereas the input variable was a function of time (T) only, after <b>classify</b> was applied the output variable became a function of both time (T) and class, which was named after the variable (avOS). So, for a given season and a given class, the output variable will have a value of either 0 (not a member of the class), 1 (is a member of the class), or NaN (missing). The Live Example Link below is followed by a link to a table from the same calculation, which will help to illustrate the meaning of the output.<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.KAPLAN/.Indices/.NINO3/.avOS/T/%28Jan%201901%29%28Dec%201990%29RANGE/T/3/boxAverage%5BT%5Dpercentileover/%7BLaNina/0.2/Neutral/0.8/ElNino%7Dclassify/">Live Example Link</a><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.KAPLAN/.Indices/.NINO3/.avOS/T/%28Jan%201901%29%28Dec%201990%29RANGE/T/3/boxAverage%5BT%5Dpercentileover/%7BLaNina/0.2/Neutral/0.8/ElNino%7Dclassify/ngridtable/3+ncoltable.html?tabopt.N=4&tabopt.1=text&tabopt.2=text&tabopt.3=text&tabopt.4=blankNaN&NaNmarker=&tabtype=html&eol=LF+%28unix%29">Table Link</a><br><br><b>classify</b> makes it very handy to calculate composites, such as long-term average seasonal precipitation conditioned upon the state of ENSO. The following example shows an application of the <b>classify</b> statement to illustrate a relationship between ENSO and monsoon rainfall in India.<br><br><font color="#008000">expert SOURCES .Indices .india .rainfall<br>SOURCES .KAPLAN .Indices .NINO3 .avOS<br>T (Oct 1901) (Dec 1990) RANGE<br>T 4 boxAverage<br>T 12 STEP<br>[T]percentileover<br>{LaNina 0.2 Neutral 0.8 ElNino}classify<br>T 4 shiftdatashort<br>[T]weighted-average<br>table: 1 :table</font><br><br>A weighted average of the June-September all-India rainfall index is taken with the classification (0 or 1) of October-January seasonal SSTAs as either 'LaNina', 'Neutral', or 'ElNino' to illustrate differences in long-term mean June-September Indian monsoon rainfall based upon the state of ENSO later in the year (an ENSO-based composite of seasonal precipitation).<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.Indices/.india/.rainfall/SOURCES/.KAPLAN/.Indices/.NINO3/.avOS/T/%28Oct%201901%29%28Dec%201990%29RANGE/T/4/boxAverage/T/12/STEP%5BT%5Dpercentileover/%7BLaNina/0.2/Neutral/0.8/ElNino%7Dclassify/T/4/shiftdatashort%5BT%5Dweighted-average/table:/1/:table/">Live Example Link<br></a><br>The following example is similar to the previous one except that instead of using a single precipitation time series for India it uses CMAP gridded precipitation values that vary in space over India. It produces composite maps of June-September 1979-2006 seasonal average precipitation according to ENSO state (LaNina, Neutral, and ElNino) over south Asia.<br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .Merged_Analysis .monthly .v0703 .ver2 .prcp_est </font><br><font color="#008000">X 60. 100. RANGEEDGES </font><br><font color="#008000">Y 0 40 RANGEEDGES </font><br><font color="#008000">T (Jun 1979) (Sep 2006) RANGE </font><br><font color="#008000">T 4 runningAverage </font><br><font color="#008000">T 12 STEP </font><br><font color="#008000">SOURCES .KAPLAN .Indices .NINO3 .avOS </font><br><font color="#008000">T (Oct 1979) (Dec 2006) RANGE </font><br><font color="#008000">T 4 boxAverage </font><br><font color="#008000">T 12 STEP </font><br><font color="#008000">[T]percentileover </font><br><font color="#008000">{LaNina 0.2 Neutral 0.8 ElNino}classify </font><br><font color="#008000">T 4 shiftdatashort </font><br><font color="#008000">[T]weighted-average</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.Merged_Analysis/.monthly/.v0703/.ver2/.prcp_est/X/60./100./RANGEEDGES/Y/0/40/RANGEEDGES/T/%28Jun%201979%29%28Sep%202006%29RANGE/T/4/runningAverage/T/12/STEP/SOURCES/.KAPLAN/.Indices/.NINO3/.avOS/T/%28Oct%201979%29%28Dec%202006%29RANGE/T/4/boxAverage/T/12/STEP%5BT%5Dpercentileover/%7BLaNina/0.2/Neutral/0.8/ElNino%7Dclassify/T/4/shiftdatashort%5BT%5Dweighted-average/">Live Example Link </a><br><br>Note that in the special case of a variable in <i>categorical form</i> (i.e. which has integral values that are chosen from a list), and that list is given with the variable, then the list of classes and transitions can be omitted. For example,<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.Merged_Analysis/.monthly/.v0703/.ver2/.prcp_est/X/60./100./RANGEEDGES/Y/0/40/RANGEEDGES/T/%28Jun%201979%29%28Sep%202006%29RANGE/T/4/runningAverage/T/12/STEP/SOURCES/.KAPLAN/.Indices/.NINO3/.avOS/T/%28Oct%201979%29%28Dec%202006%29RANGE/T/4/boxAverage/T/12/STEP%5BT%5Dpercentileover/%7BLaNina/0.2/Neutral/0.8/ElNino%7Dclassify/T/4/shiftdatashort%5BT%5Dweighted-average/"> </a><font color="#008000">SOURCES .NASA .ISLSCP .GDSLAM .Hydrology-Soils .soils .texture classify </font><br><br>transforms the variable from <i>categorical form </i>(a 3D variable of integer values) to <i>complete disjunctive form</i> (a 4D variable of weights (0 or 1) where the added dimension has the list of possibilities defined with the original categorical dataset). This dataset can now be regridded or factor-analyzed.<font color="#008000"> </font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NASA/.ISLSCP/.GDSLAM/.Hydrology-Soils/.soils/.texture/classify/">Live Example Link</a> Classifies data into categories, i.e. labels ranges of values. 2 classes true alternating names and numbers, starting and ending with a name, so that there are N+1 names and N numbers categorical form 3 facet true name of new independent variable (name of var if omitted) Gives the distributions of values for a given variable d(x,y,z,t) by returning a new variable f(d) which gives the number of occurrences of data points nearest the values d_i which are equally spaced from min to max in steps of step distrib f(d) 2 Variable returning the number of occurrences of data points nearest the values d_i which are equally spaced from min to max in steps of step false distrib2 1 d (x,y,z,t) false Variable to be used distrib1 gives the distributions of values for a given variable d(x,y,z,t) by returning a new variable f(d) which gives the number of occurrences of data points nearest the values d_i which are equally spaced from min to max in steps of step. distrib distrib2D 3 C distrib2D3 false Bivariate counts of the variables above 2 false B distrib2D2 Variable to be processed 1 Variable to be processed A distrib2D1 false distrib2D returns bivariate counts Returns bivariate counts width96 computes the 96 percentile widths Computes the 96 percentile widths var' 2 96 percentile widths width962 false width961 var 1 Variable the computation will be derived from false width96 Returns the frequency distribution of a set of data for a specified range and step interval. Commonly used to create histograms. upper upper bound of range over which distribution is to be found 4 false false dist frequency distribution of <i>var</i> within range and interval specified by <i>lower</i>, <i>upper</i>, and <i>step</i> 8 Returns the frequency distribution of a set of data for a specified range and step interval. Commonly used to create histograms. <br> <u><b>Description</b></u><br><br><b>distrib1D</b> returns the frequency distribution (as binned counts) of data from an input variable based upon a user-specified binning interval and range limits defined in the DATA lower upper step RANGESTEP command. In doing this, <b>distrib1D</b> creates a new grid of bins defined by the RANGESTEP command that has the same name as the input variable.<br><br><u><b>Example</b></u><br><br>In this example soil moisture values from July 1948-2003 and the grid box that includes the coordinates 20&#176; N, 50&#176; E have been selected, leaving soil moisture values that vary only over the time grid. <i>DATA 0 50 2 RANGESTEP </i>specifies a selection of soil moisture values from 0 to 50 mm and a binning of 2 mm. <b>distrib1D</b> uses these specifications to produce a count of the number of soil moisture observations in the selection that fall within each 2 mm bin from 0 to 50 mm. The limits of the RANGESTEP command serve as bin centers. Therefore, in this example, there are 26 bins, namely -1 (actually 0)-1, 1-3, 3-5, ..., 47-49, 49-51 (actually 50). The grid of bins is named &quot;w&quot;.<br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .GMSM .w<br>Y (20) VALUE<br>X (50) VALUE<br>T (Jul 1948-2003) VALUES<br>DATA 0 50 2 RANGESTEP<br>distrib1D</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.GMSM/.w/Y/%2820%29VALUE/X/%2850%29VALUE/T/%28Jul%201948-2003%29VALUES/DATA/0/50/2/RANGESTEP/distrib1D/">Live Example Link</a><br><br>In the following example using the GHCN monthly precipitation data set, precipitation values for stations in Madagascar and its vicinity have been selected for January 1971 to December 2000. The <i>DATA 0 1200 50 RANGESTEP</i> command specifies the selection of precipitation values from 0 to 1200 mm and defines 50 mm-wide bins. Finally, in the <b>distrib1D</b> command, the inclusion of IWMO, the name of the station ID grid, in brackets means that a frequency distribution will be constructed for each station separately. If &quot;[IWMO]&quot; had been excluded, a single frequency distribution would have been produced using the collective precipitation values from all the selected stations. This can be generalized to multiple grids (if the data depend upon additional grids) as well.<br><br><font color="#008000">SOURCES .NOAA .NCDC .GHCN .v2beta<br>lon (41.5) (53.0) masknotrange SELECT<br>lat (-28.5) (-8.0) masknotrange SELECT<br>T (Jan 1971) (Dec 2000) RANGE<br>.prcp<br>DATA 0 1200 50 RANGESTEP<br>[IWMO]distrib1D</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.GHCN/.v2beta/lon/%2841.5%29%2853.0%29masknotrange/SELECT/lat/%28-28.5%29%28-8.0%29masknotrange/SELECT/T/%28Jan%201971%29%28Dec%202000%29RANGE/.prcp/DATA/0/1200/50/RANGESTEP%5BIWMO%5Ddistrib1D/">Live Example Link</a><br><br> 5 width of intervals (in units of <i>var</i>) used in distribution false step DATA 2 false RANGESTEP false 6 3 false lower bound of range over which distribution is to be found lower depends_on independent variables that the distribution is to depend on. Default is to compute over all independent variables, thus depends_on is empty true 7 false var variable (i.e., data) of which distribution is to be found 1 distrib1D Categorization input data to be classified 1 false var 1 variable of weights in complete disjunctive form weights false returns class with the highest weight dominant_class returns class with the highest weight <b>dominant_class </b>transforms the variable from <i>complete disjunctive form</i> (a variable of weights (0 to 1) with an independent variable that corresponds to the list of classes) to <i>categorical form </i>(a variable of integer values, numbering the classes from 1 to N) by choosing the class with the highest non-zero weight (first class is used in the case of multiple classes with the same non-zero weight). When multiple independent variables are specified, dominant_class combines the multiple class sets, creating the multivariate class that corresponds to the combination of the class sets. false one or more independent variables which list the classes 2 classlist var false data of which monthly climatology is to be calculated <p>Note that the data must be monthly (i.e., units of time grid must be months). 1 climovar monthly climatology of <i>var</i> based on all selected years. <i>climovar</i> is dependent on a time grid with 12 grid points (i.e., Jan, Feb, ..., Nov, Dec). false 2 yearly-climatology SOURCES .UNH .CSRC .RivDIS .dischrg <br>ISTA 518 VALUE <br>T (Jan 1930) (Dec 1979) RANGE <br>yearly-climatology Calculates a monthly climatology Calculates a monthly climatology 6 log Result 2 <i>Result</i> will be of the same type as <i>A</i> (i.e., either a variable or constant). 1 A variable or constant of which the log (base 10) is taken Calculates the logarithm of a number or variable, with base 10 log Calculates the logarithm of a number or variable, with base 10 density Unecso '81 density (insitu) of seawater minus 1 false densa4 4 <p style="margin-top: 0"> <br> <b><u>Description</u></b><br><br><b>densa</b> calculates the density of seawater above 1 g/cm³ given input variables temperature (T), salinity (S), and pressure (P), in that order, using the international equation of state for seawater, IES80 (UNESCO 1981).<br><br><b><u>Reference</u></b><br><br>UNESCO, 1981: Tenth report of the joint panel on oceanographic tables and standards. UNESCO Tech. Paper in Marine Science 36, 25pp.<br><br><b><u>Example</u></b><br><br>The following example calculates the density of seawater based on samples of measured temperature, salinity, and pressure from the full data set of GEOSECS observational profiles.<br><br><font color="#008000">SOURCES .GEOSECS .TEMP </font> </p> <p style="margin-top: 0"> <font color="#008000">SOURCES .GEOSECS .SAL </font> </p> <p style="margin-top: 0"> <font color="#008000">SOURCES .GEOSECS .PRESS </font> </p> <p style="margin-top: 0"> <font color="#008000">densa</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.GEOSECS/.TEMP/SOURCES/.GEOSECS/.SAL/SOURCES/.GEOSECS/.PRESS/densa/">Live Example Link</a> </p> S false 2 Salinity densa2 densa 3 false Pressure densa3 P densa false 1 T Temperature densa1 unecso '81 density (insitu) of seawater minus 1 Unecso '81 density (insitu) of seawater minus 1 products ivar products2 2 Independent variable along which function will be applied false 1 var Variable to which function will be applied to false products1 3 products3 false var' Pairwise products along grid of variable takes pairwise products along independent variable of variable products Takes pairwise products along independent variable of variable T pressure1 false 1 Temperature S pressure2 2 Salinity false Computes hydrostatic pressure by integrating from the surface pressure3 Height false 3 Z computes hydrostatic pressure by integrating from the surface pressure Hydrostatic pressure P false 4 pressure4 pressure dailyvar daily data to be averaged, with units of units false 1 minfrac true Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within each year in order for each average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the average is calculated regardless of the amount of data present. 2 yearlyvar 3 yearly average of dailyvar false seasonalAverage false data input dataset or variable 1 computes seasonal Average over time computes seasonal Average over time independent variable that represents time 2 false <p style="margin-top: 0"> seasonalAverage can be used to pick out a season from data that depends on time. </p> <p style="margin-top: 0"> For example, consider </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> SOURCES .NOAA .NCEP .EMC .CMB .GLOBAL .Reyn_SmithOIv2 .weekly .ssta </p> <p style="margin-top: 0"> T (Nov-Jan) seasonalAverage </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> This is also accessible <a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.EMC/.CMB/.GLOBAL/.Reyn_SmithOIv2/.weekly/.ssta/T/%28Nov-Jan%29seasonalAverage/">here</a>. </p> 5 seasonally averaged data dataset or variable that has been seasonally averaged false 3 season length false true minfrac 4 converts daily data to yearly by averaging converts daily data to yearly by averaging yearlyAverage <br> <b><u>Description</u></b><br><br><b>abs</b> applies sqrt[<i>num</i>²] to a variable stream of real numbers or a constant (<i>num</i>) to produce its numerical magnitude without regard to its sign.<br><br><b><u>Example</u></b><br><br><font color="#008000">SOURCES .NOAA .NCEP-NCAR .CDAS-1 .MONTHLY .Diagnostic .above_ground .u<br>abs </font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.MONTHLY/.Diagnostic/.above_ground/.u/abs/">Live Example Link</a><br> num false variable or constant of which the absolute value is to be found 1 absnum false 2 absolute value of <i>num</i> Returns the absolute value of a variable or a constant Returns the absolute value of a variable or a constant abs Samples variable along ivar1 according to ivar2, returning lowest value in ivar1 greater than or equal to value in ivar2 SAMPLELB 2 Independent variable from which variable is sampled along ivar1 false Variable to be sampled false 1 var 4 Variable with lowest value in ivar1 greater than or equal to value in ivar2 false var' Independent variable from which ivar1 is based 3 false ivar2 takes pairwise ratios along independent variable of variable false ratios2 Independent variable along which function will be applied ivar 2 ratios1 false Variable to which function will be applied 1 var false stream' ratios3 3 Pairwise ratios along grid of variable Takes pairwise ratios along independent variable of variable ratios ratios false 3 monthlyvar monthly average of <i>dailyvar</i> <p> <i>monthlyvar</i> now has a temporal resolution of months. Note: for variables with units of accumulation (e.g., precipitation), the units are now <i>units</i>/day. This should be changed in the interface as shown in the example below. daily data to be averaged, with units of <i>units</i> false 1 dailyvar Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within each month in order for each monthly average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the average is calculated regardless of the amount of data present. true minfrac 2 SOURCES .NOAA .NCEP .CPC .FEWS .DAILY .est_prcp <br>0.9 monthlyAverage <br>/units (mm/day) def <p> OR <p> SOURCES .NOAA .NCDC .GDCN .TMAX <br>ISTA 253175 VALUE <br>monthlyAverage Converts daily data to monthly data by averaging monthlyAverage Converts daily data to monthly data by averaging returns a variable of random numbers randomdata randomdata2 2 var' Random numbers false 1 Variable to be converted false randomdata1 var randomdata Returns a variable of random numbers false var 1 variable (i.e., data) of which distribution is to be found dist cumulative frequency distribution of <i>var</i> within range and intervals specified by <i>lower</i>, <i>upper</i>, and <i>step</i> false 7 width of intervals (in units of <i>var</i>) used to find distribution false step 5 4 false upper upper bound of range over which distribution is to be found SOURCES .NOAA .NCEP .CPC .GMSM .w <br>Y (20) VALUE <br>X (50) VALUE <br>T (Jul 1948-2003) VALUES <br>DATA 0 50 0.5 RANGESTEP <br>integrateddistrib1D integrateddistrib1D 2 DATA false false 6 RANGESTEP false lower bound of range over which distribution is to be found lower 3 Returns the cumulative frequency distribution of a set of data for a specified range and step interval. Returns the cumulative frequency distribution of a set of data for a specified range and step interval. dataset or variable (<i>ds/var</i>) restricted to only data associated with <i>grid</i> values within <i>range_low</i> and <i>range_high</i>, inclusive; applied to all variables in the dataset if no variable selected restricted_ds/var false 5 grid in which values will be selected false grid 2 SOURCES .CAC <br>Y (10S) (10N) RANGEEDGES <p> OR <p> SOURCES .CAC .ssta <br>Y -10 10 RANGEEDGES upper threshold of range (upper edge of closest grid box) range_high 4 false removeVALUES removes list of VALUES from variable (complement to VALUES) false data_out 4 new variable without listed values value list 3 values to be dropped from set false false ivar independent variable along which selection is to be performed 2 1 data_variable variable on which selection is to be performed false removes list of VALUES from variable (complement to VALUES) false dataset or variable dependent on the grid that will be sampled along ds/var 1 Identifies continuous range of values in one grid (i.e., independent variable) for which data will be selected. RANGEEDGES false lower threshold of range (lower edge of closest grid box) range_low 3 Identifies continuous range of values in one grid (i.e., independent variable) for which data will be selected. 1 dataset or variable dependent on the grid that will be sampled along ds/var false Identifies continuous range of values in one grid (i.e., independent variable) for which data will be selected. false restricted_ds/var 5 dataset or variable (<i>ds/var</i>) restricted to only data associated with <i>grid</i> values within <i>range_low</i> and <i>range_high</i>, inclusive; applied to all variables in the dataset if no variable selected RANGE SOURCES .CAC <br> Y (10S) (10N) RANGE <p>OR<p> SOURCES .CAC .ssta<br> Y -10 10 RANGE 3 lower threshold of range (center of closest grid box) range_low Identifies continuous range of values in one grid (i.e., independent variable) for which data will be selected. 4 range_high upper threshold of range (center of closest grid box) grid 2 grid in which values will be selected 1 var data of which percentiles are to be found false Replaces data values with their percentile, based on non-parametric methods. grid(s) over which percentiles are to be found 2 false grid SOURCES .UNH .CSRC .RivDIS .dischrg <br>ISTA 4 VALUE <br>T (Aug-Oct 1950-1979) RANGE <br>[T]percentileover 3 true minfrac Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the percentiles to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the percentiles are calculated regardless of the amount of data present in domain. Replaces data values with their percentile, based on non-parametric methods. percentileover 4 percentiles of <i>var</i> over <i>grid</i>. <i>percentilevar</i> has same name and is dependent on the same grids as <i>var</i>. percentilevar false false data to be analyzed var 0 true weights wghts 1 3 false time grids over which svd is to be found time svd Computes singular value decomposition 2 false spatial grids over which svd is to be found space dataset comprised of the singular value decomposition of <i>var</i>, where structures (Ss), time series (Ts), singular values (sv), and normalized eigenvalues (evaln) are variables. svd_dataset 4 N 2 false number of modes to include in rotation svddataset dataset of svd results false 1 dataset containing the rotated eofs rotated_eofs 3 false performs varimax rotation on the results of svd performs varimax rotation on the results of svd varimax Computes singular value decomposition SOURCES .CAC .ssta <br>{ Y cosd } [ X Y ] [T] svd <p> As a check of normalization, plot<p> Ss dup mul Y cosd mul [X Y]average <p> or <p> Ts sv div dup mul [T]average <p> Both should be a constant 1. rasterizes a geometry to match a gridded variable rasterizes a geometry to match a gridded variable 2 geometry variable which is to be rasterized. false geometry geometry gridded to match griddedvar griddedgeometry false 3 griddedvar false 1 reference gridded variable to be matched false rectangular geometry variable 5 rectangle <p style="margin-top: 0"> <br> <u><b>Description</b></u><br><br><b>rect</b> defines a rectangular region from user-defined coordinates within an input variable, creating a geometry variable &quot;<i>rectangle</i>&quot; as a result. To define the rectangle, the user provides four arguments in the following order: <i>xlo</i>, <i>ylo</i>, <i>xhi</i>, and <i>yhi.</i><br><br> </p> false 2 low value of second coordinate ylo defines a rectangle geometry rect defines a rectangle geometry 3 xhi high value of first coordinate false low value of first coordinate 1 false xlo yhi high value of second coordinate 4 false rasterize SM121 4 n number of passes of smoothing to be applied false Applies two-dimensional 121 smoothing to a variable grids 2 false grids (1 or 2) over which 121 smoothing is to be applied SOURCES .NOAA .NCEP .CPC .CAMS_OPI .climatology .prcp <br>[X Y] 1 SM121 false 1 var variable to be smoothed Applies two-dimensional 121 smoothing to a variable 5 false smoothvar <i>var</i> after <i>n</i> passes of 121 smoothing has been applied over <i>grids</i> definite integral over <i>ivar</i> from <i>low</i> to <i>high</i> integral 4 Returns integral evaluated between two limits Returns integral evaluated between two limits definite-integral 0 integrand function to be integrated independent variable to be integrated over ivar 1 2 low lower limit <br> <u><b>Description</b></u><br><br><b>definite-integral</b> calculates the definite integral of the latest variable on the stack over the grid (independent variable) and limits (<i>low</i> and <i>high</i>) specified in the command.<br><br><u><b>Example</b></u><br><br><font color="#008000">SOURCES .NOAA .NCEP-NCAR .CDAS-1 .DAILY .Intrinsic .PressureLevel .reld<br>T (1 Jan 2007) VALUE<br>P 600 10 definite-integral</font><br><br>In this example global gridded relative divergence values for 1 January 2007 are selected for all available vertical pressure levels from the NCEP/NCAR Reanalysis data set. Then, the integral of the relative divergence values is taken over the vertical pressure grid (P) from 600 to 10 hectopascals (mb), presumably from near the level of non-divergence to the top of the atmosphere in the Reanalysis data set. To the extent that the relative divergence values are accurate, the result of the calculation should be proportional to the vertical velocity at 600 hPa. Maps of the result of this calculation and the Reanalysis pressure vertical velocity for the same day are given here:<br><br> <table border="0" cellpadding="1" cellspacing="1"> <tr> <td align="center"> Vertical Integral of Relative Divergence<br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.reld/T/%281%20Jan%202007%29VALUE/P/600/10/definite-integral/figviewer.html?my.help=more+options&map.Y.units=degree_north&map.Y.plotlast=90N&map.url=X+Y+fig-+colors+coasts+lakes+-fig&map.domain=+%7B+Y+-90.+90.+plotrange+%7D&map.domainparam=+%2Fplotaxislength+432+psdef+%2Fplotborder+72+psdef&map.zoom=Zoom&redraw.x=22&redraw.y=9&map.Y.plotfirst=90S&map.X.plotfirst=1.25W&map.X.units=degree_east&map.X.modulus=360&map.X.plotlast=1.25W&map.int_dP.plotfirst=-0.0051178&map.int_dP.units=100+kilogram+meter-1+second-3&map.int_dP.plotlast=0.00511725&map.newurl.grid0=X&map.newurl.grid1=Y&map.newurl.land=draw+coasts&map.newurl.plot=colors&map.plotaxislength=300&map.plotborder=72&map.fnt=Helvetica&map.fntsze=12&map.XOVY=auto&map.color_smoothing=1&map.framelbl=framelabelstart&map.framelabeltext=&map.iftime=25&map.mftime=25&map.fftime=200"><img src="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.reld/T/%281%20Jan%202007%29VALUE/P/600/10/definite-integral/X+Y+fig-+colors+coasts+-fig+//int_dP/-0.0051178/0.00511725/plotrange/Y/-90/90/plotrange//plotaxislength+300+psdef//plotborder+72+psdef//XOVY+null+psdef+.gif"></a><br><img src="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.reld/T/%281%20Jan%202007%29VALUE/P/600/10/definite-integral/X+Y+fig-+colors+coasts+-fig+.auxfig+//int_dP/-0.0051178/0.00511725/plotrange/Y/-90/90/plotrange//plotaxislength+300+psdef//plotborder+72+psdef//XOVY+null+psdef+.gif"> </td> <td align="center"> Pressure Vertical Velocity at 600 hPa<br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.vvel/P/600/VALUE/T/%281%20Jan%202007%29VALUE/figviewer.html?my.help=more+options&map.Y.units=degree_north&map.Y.plotlast=90N&map.url=X+Y+fig-+colors+coasts+-fig&map.domain=+%7B+%2Fvvel+-0.6+0.6+plotrange+Y+-90+90+plotrange+%7D&map.domainparam=+%2Fplotaxislength+432+psdef+%2Fplotborder+72+psdef+%2FXOVY+null+psdef&map.zoom=Zoom&redraw.x=24&redraw.y=-3&map.Y.plotfirst=90S&map.X.plotfirst=1.25W&map.X.units=degree_east&map.X.modulus=360&map.X.plotlast=1.25W&map.vvel.plotfirst=-0.6&map.vvel.units=Pa%2Fs&map.vvel.plotlast=0.6&map.newurl.grid0=X&map.newurl.grid1=Y&map.newurl.land=draw+coasts&map.newurl.plot=colors&map.plotaxislength=300&map.plotborder=72&map.fnt=Helvetica&map.fntsze=12&map.XOVY=auto&map.color_smoothing=1&map.framelbl=framelabelstart&map.framelabeltext=&map.iftime=25&map.mftime=25&map.fftime=200"><img src="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.vvel/P/600/VALUE/T/%281%20Jan%202007%29VALUE/X+Y+fig-+colors+coasts+-fig+//vvel/-0.6/0.6/plotrange/Y/-90/90/plotrange//plotaxislength+300+psdef//plotborder+72+psdef//XOVY+null+psdef+.gif"></a><br><img src="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.vvel/P/600/VALUE/T/%281%20Jan%202007%29VALUE/X+Y+fig-+colors+coasts+-fig+.auxfig+//vvel/-0.6/0.6/plotrange/Y/-90/90/plotrange//plotaxislength+300+psdef//plotborder+72+psdef//XOVY+null+psdef+.gif"> </td> </tr> </table> <p> <a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP-NCAR/.CDAS-1/.DAILY/.Intrinsic/.PressureLevel/.reld/T/%281%20Jan%202007%29VALUE/P/600/10/definite-integral/">Live Example Link</a> </p> <p> </p> <br> upper limit high 3 false var 1 variable dependent on grid to be split splitstreamgrid Splits grid (i.e., independent variable) into two grids. Commonly-used to create climatologies (see live example below). false step interval on which grid split is based 3 step 4 false varout same as <i>var</i> except now dependent on two newly-defined grids: <br>1) <i>grid_in</i> now contains <i>step</i> points. In the example below, <i>grid_in</i> (i.e., T) contains values of Jan, Feb, ..., Dec. <br> 2) <i>newgrid</i> contains <i>n</i>/<i>step</i> points with a step interval of <i>step</i>. In the example below, <i>newgrid</i> (i.e., T2) contains values of 1982, 1983, ..., 2003. Splits grid (i.e., independent variable) into two grids. Commonly-used to create climatologies (see live example below). <u><b>Description</b></u><br><b>splitstreamgrid</b> splits grid (i.e., independent variable) into two grids. Commonly used to create climatologies.<br><u><b>Example</b></u><br><font color="#008000">SOURCES .NOAA .NCEP .EMC .CMB .GLOBAL .Reyn_SmithOIv2 .monthly .sst<br>T (Jan 1982) (Dec 2003) RANGE<br>T 12 splitstreamgrid<br></font><font color="#800000"><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.EMC/.CMB/.GLOBAL/.Reyn_SmithOIv2/.monthly/.sst/T/%28Jan%201982%29%28Dec%202003%29RANGE/T/12/splitstreamgrid%5BT2%5Daverage/">Live Example Link</a></font> 2 grid_in false grid to be split, containing <i>n</i> points false 4 Variable1 regridded to match variable2 var1' Name of object object name 3 false false var2 2 Variable to which variable1 will be regridded to false 1 var1 Variable to be regridded gridtomatchnamed false mulavg2 2 Variable to be multiplied and averaged var2 Multiplies and averages over a set of independent variables Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the average is calculated regardless of the amount of data present in domain 4 minfrac true mulavg4 mulavg3 3 Grids over which variables will be multiplied and averaged ivar false mulavg multiplies and averages over a set of independent variables 5 false mulavg5 var' Variable multiplied and avereaged over a set of grids mulavg 1 false Variable to be multiplied and averaged mulavg1 var1 creates a CPT v10 file from the input dataset/variable 5 false cpt file page which gives links to the CPT files output file false grid or variable that gives the time 4 time 3 spatial grids either a pair of spatial grids or the station grid false true 2 optional list of variables to be extracted from the dataset (default is to include all of them) variable list 1 false inputdata dataset or variable to be put in cptv10 file <p style="margin-top: 0"> <b><font face="SansSerif">cptv10 </font></b>allows one to explicitly specify variables, spatial, and temporal grids for output to a cptv10 file. Since CPT is most useful as a tool to analyze results from Data Library analyses (as opposed to direct downloads of datasets that explicitly exist in the data library), the flexibility of explicitly specifying information for CPT is useful. </p> <h4> Downloading a precipitation analysis with no explicit time </h4> <pre> SOURCES .IRI .FD .ECHAM4p5 .Forecast .ca_sst .ensemble24 .MONTHLY .prec (mm/day) unitconvert [M]average S (1 Dec 1957-2008) VALUES L 0.5 2.5 RANGEEDGES [L] /keepgrids average [X Y][S L add]cptv10 </pre> This example accesses the precipitation record, changes the units, ensemble averages, selects Dec starts, and averages over the first three lags (retaining the lag (L) grid so that it can be used to compute time later). The last line then specifies X Y as the spatial grids and the sum of start and lead time as time. <p> <a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.IRI/.FD/.ECHAM4p5/.Forecast/.ca_sst/.ensemble24/.MONTHLY/.prec/(mm/day)unitconvert%5BM%5Daverage/S/(1%20Dec%201957-2008)VALUES/L/0.5/2.5/RANGEEDGES%5BL%5D//keepgrids/average%5BX/Y%5D%5BS/L/add%5Dcptv10/">Results</a> <h4> Downloading two precipitation analyses as a pair of fields for CPT </h4> <pre> SOURCES .NOAA .NCEP .EMC .CFS .MONTHLY .surface .prate L (1.5) (3.5) RANGEEDGES S (0000 1 Sep 1982-2008) VALUES X (30E) (100E) RANGEEDGES Y (30N) (30S) RANGEEDGES [M]average [L] /keepgrids average c: /name /water_density def 998 (kg/m3) :c div (mm/day) unitconvert SOURCES .IRI .MP .RESEARCH .COUPLED .GLOBAL .ECHAM4p5-MOM3-realtime .ATM .Surface .prec L (1.5) (3.5) RANGEEDGES S (0000 1 Sep 1982-2008) VALUES X (30E) (100E) RANGEEDGES Y (30N) (30S) RANGEEDGES [M]average [L] /keepgrids average (mm/day) unitconvert [ X Y ] regridLinear {NOAA_prcp IRI_prcp}ds [X Y][S L add ]cptv10</pre> In this example, we compute seasonal average precipitation from two different models. Extract data from the monthly CFS, Sep starts averaged from 1.5 to 3.5 month leads. Note that we keep the L grid in the average so that we can use it later to compute time from start time and lead. <pre> SOURCES .NOAA .NCEP .EMC .CFS .MONTHLY .surface .prate L (1.5) (3.5) RANGEEDGES S (0000 1 Sep 1982-2008) VALUES X (30E) (100E) RANGEEDGES Y (30N) (30S) RANGEEDGES [M]average [L] /keepgrids average </pre> Because this model outputs precipitation in kg /m2 /s, we need to divide by water density to allow unit conversion to mm/day. <pre> c: /name /water_density def 998 (kg/m3) :c div (mm/day) unitconvert </pre> Now we extract data from the monthly MOM3, again Sep starts averaged from 1.5 to 3.5 month leads. The mm/day unit conversion does not involve water density this time. <pre> SOURCES .IRI .MP .RESEARCH .COUPLED .GLOBAL .ECHAM4p5-MOM3-realtime .ATM .Surface .prec L (1.5) (3.5) RANGEEDGES S (0000 1 Sep 1982-2008) VALUES X (30E) (100E) RANGEEDGES Y (30N) (30S) RANGEEDGES [M]average [L] /keepgrids average (mm/day) unitconvert </pre> Now we regrid the IRI model analysis to match the NOAA model analysis <pre> [ X Y ] regridLinear </pre> We bundle the two results into a new dataset, calling them NOAA_prcp and IRI_prcp respectively. <pre> {NOAA_prcp IRI_prcp}ds </pre> We then invoke cptv10, specifying X Y spatial grids, and computing time from the sum of start time S and lead time L. Note that the width of three months associated with L is inherited by T, so that the output is correctly labelled as seasonal averages. <pre> [X Y][S L add ]cptv10 </pre> <a href="http://iridl.ldeo.columbia.edu/SOURCES/.NOAA/.NCEP/.EMC/.CFS/.MONTHLY/.surface/.prate/L/(1.5)(3.5)RANGEEDGES/S/(0000%201%20Sep%201982-2008)VALUES/X/(30E)(100E)RANGEEDGES/Y/(30N)(30S)RANGEEDGES%5BM%5Daverage%5BL%5D//keepgrids/average/c://name//water_density/def/998/(kg/m3):c/div/(mm/day)unitconvert/SOURCES/.IRI/.MP/.RESEARCH/.COUPLED/.GLOBAL/.ECHAM4p5-MOM3-realtime/.ATM/.Surface/.prec/L/(1.5)(3.5)RANGEEDGES/S/(0000%201%20Sep%201982-2008)VALUES/X/(30E)(100E)RANGEEDGES/Y/(30N)(30S)RANGEEDGES%5BM%5Daverage%5BL%5D//keepgrids/average/(mm/day)unitconvert%5BX/Y%5DregridLinear/%7BNOAA_prcp/IRI_prcp%7Dds%5BX/Y%5D%5BS/L/add%5Dcptv10/">Results</a><a> <h4>Downloading dual field Station data</h4> <pre> SOURCES .NOAA .NCDC .USHCN state (30) dup masknotrange SELECT T (Jun 1960-1979) VALUES {raw .prcp raw .mean .temp lon lat elev Name}[ID][T] cptv10 </pre> This example uses the state code to pick out a subset of the stations, picks June data from 1960-1979, then selected prcp and temp data, with lon, lat, elev, and Name station information. <p> <a href="http://iridl.ldeo.columbia.edu/SOURCES/.NOAA/.NCDC/.USHCN/state/(30)dup/masknotrange/SELECT/T/(Jun 1960-1979)VALUES/{raw/.prcp/raw/.mean/.temp/lon/lat/elev/Name}[ID][T]/cptv10/">Results</a> cptv10 SELECT Selects values of a grid based on values of the indicated variable. Commonly used to select stations - station ids (grid) selected based on lat/lon (variables) values (see Live Example). Selects values of a grid based on values of the indicated variable. Commonly used to select stations - station ids (grid) selected based on lat/lon (variables) values (see Live Example). <u><b>Description </b></u><br><b>SELECT</b> is used to select values of a grid based on values of the indicated variable. Commonly used to select stations - station ids (grid) selected based on lat/lon (variables) values<br><u><b>Examples</b></u><br><font color="#008000">lat SELECT</font><br><font color="#800000"><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.GDCN/lon/290/340/masknotrange/SELECT/lat/-10/10/masknotrange/SELECT/">Live Example Link</a></font><br> false 3 selected ds new dataset selected from <i>ds</i>. All the values of the independent variable of <i>var</i> that are marked missing in <i>var</i> are eliminated false 1 ds dataset to be selected from var variable used to make selection of grid values 2 false var' Partial derivative of stream along grid 3 partial3 false ivar partial2 Independent variable along which partial derivative of variable will be taken 2 false partial1 Variable of which partial derivative will be taken along grid 1 var partial partial takes partial derivative of variable along grid Takes partial derivative of variable along grid Returns the minimum value of a variable over a selected grid(s) SOURCES .NOAA .NCEP .CPC .CAMS .station .temperature .temp <br>IWMO (1001) VALUES <br>[T]0.9 minover <p> OR <p> SOURCES .NOAA .NCEP .CPC .GMSM .w <br>X 112 153 RANGE <br>Y -44 -11 RANGE <br>[X Y]minover grid(s) over which minimum value is found 2 false grid variable of which minimum value is found false 1 var minimum value of <i>var</i> within selected domain of <i>grid</i> <p> <i>minvar/num</i> is no longer dependendent on <i>grid</i>, but is still dependent on any other grids that <i>var</i> depended on (if any) minvar/num false 4 3 true Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within theselected domain in order for the minimum value to be found. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the minimum value is found regardless of the amount of data present in domain. minfrac minover Returns the minimum value of a variable over a selected grid(s) beta4 false 4 beta beta beta3 false Pressure P 3 beta2 Salinity S 2 false beta <br> <u><b>Description</b></u><br><br><b>beta</b> uses potential temperature, salinity, and pressure values in the ocean to calculate beta (the saline contraction coefficient).<br><br><u><b>References</b></u><br><br>McDougall, T. J., 1987: Neutral Surfaces. <i>Journal of Physical Oceanography</i>, <b>17</b>, 1950-1964.<br><br><u><b>Example</b></u><br><br><font color="#008000">SOURCES .GEOSECS .THETA<br>SOURCES .GEOSECS .SAL<br>SOURCES .GEOSECS .PRESS<br>beta</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.GEOSECS/.THETA/SOURCES/.GEOSECS/.SAL/SOURCES/.GEOSECS/.PRESS/beta/">Live Example Link</a><br> false theta beta1 1 Potential Temperature saline contraction coefficient in the ocean 2 grid (i.e., independent variable) to be removed <p> Must contain only a single value, <i>val</i>. false grid SOURCES .NOAA .NODC .WOA01 .Grid-1x1 .Annual .an .aou <br>Z (0.0) VALUE <br>Z removeGRID Removes a single-valued grid (i.e. independent variable) from a variable false 1 var variable dependent on grid to be removed removeGRID Removes a single-valued grid (i.e. independent variable) from a variable <i>var</i> at <i>grid</i> value <i>val</i> only <p> <i>var_at_val</i> is no longer explicitly dependent on <i>grid</i>. 3 false var_at_val changes second variable to match truncation of the first Changes second variable to match truncation of the first matchtruncation matchtruncation Variable to be compared to var1 matchtruncation1 false 1 2 Variable to be truncated to match the first var2 false matchtruncation2 false matchtruncation4 4 var2' Variable truncated to match the first 3 Unchanged variable var1 matchtruncation3 false false 4 Ratio of alpha to beta abrat4 alpha/beta abrat Pressure P abrat3 3 false <br> <u><b>Description</b></u><br><br><b>abrat</b> uses potential temperature, salinity, and pressure values in the ocean to calculate the ratio of alpha (the thermal expansion coefficient) to beta (the saline contraction coefficient).<br><br><u><b>References</b></u><br><br>McDougall, T. J., 1987: Neutral Surfaces. <i>Journal of Physical Oceanography</i>, <b>17</b>, 1950-1964.<br><br><u><b>Example</b></u><br><br><font color="#008000">SOURCES .GEOSECS .THETA<br>SOURCES .GEOSECS .SAL<br>SOURCES .GEOSECS .PRESS<br>abrat</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.GEOSECS/.THETA/SOURCES/.GEOSECS/.SAL/SOURCES/.GEOSECS/.PRESS/abrat/">Live Example Link</a><br> Salinity abrat2 2 false S abrat1 theta false 1 Potential Temperature abrat ratio of alpha to beta (oceanic eq. of state) Ratio of alpha to beta avgvar false 5 running average of <i>var</i>, where the overlapping intervals are of length <i>interval</i> SOURCES .NOAA .NCDC .GHCN .v2beta .prcp <br>IWMO 63450000 VALUE <br>T 12 runningAverage runningAverage 2 grid grid(s) over which the running average is to be calculated false Calculates the running average 3 width (in units of <i>grid</i>) of the overlapping interval interval false var 1 false variable to be averaged Calculates the running average true minfrac 4 Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the running average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the running average is calculated regardless of the amount of data present in domain. CofA=B-bounded3 3 false C Converts a variable A to a variable B using a table B(C) and linear interpolation. Out of range values are pegged to the extreme values false C' CofA=B-bounded4 4 Converted variable using table B(C) and linear interpolation Out of range values beyond half a grid step are pegged to the extreme values <p style="margin-top: 0"> <br> <b><u>Description</u></b><br><br><b>CofA=B-bounded</b> accepts two variables (A and B) as input and converts variable A values to values expressed in terms of variable B based upon a table of corresponding values (B(C)) in variable B and linear interpolation between the points specified in the table. Out-of-range values are pegged to the extreme values.<br><br><b><u>Example</u></b><br><br>In the following example the input variable A is CAMS-OPI gridded monthly precipitation, which depends upon longitude (X), latitude (Y), and months (T) from Jan 1979 to Dec 2000. Input variable B is gridded observed monthly precipitation, which depends upon grids X and Y (which match with X and Y from variable A) and months (T) specified from Jan 1969 to Dec 1998. The purpose of this particular script is to use a specific 30-year set of precipitation values (variable B) to calculate precipitation percentiles by month of year and then express precipitation values from the shorter time series (variable A) in terms of the percentiles from variable B's climatology at each (X,Y) grid point. </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> Here, the time grid for variable B is split into a grid of 12 months per year (T) and a grid of 30 years (T2) from 1969 to 1998. The function &quot;replacebypercentile&quot; is applied to the T2 grid of years in variable B, with the result that the T2 grid is replaced by a new grid named &quot;percentile&quot; that has points at 0.0, 0.33, 0.5, 0.67, and 1, containing the precipitation values that correspond to those percentiles in the 30-year climatology for each month of the year at each (X,Y) grid point. </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> Based upon the correspondence between percentile and precipitation values in variable B function (the table B(C)) &quot;[percentile]CofA=B-bounded&quot; then converts the time series of monthly precipitation values in variable A to percentiles by month of year using linear interpolation between the percentile points specified in the &quot;percentile&quot; grid in variable B. The word &quot;bounded&quot; in the name of the function indicates that any precipitation values in variable A that fall outside the range of precipitation values that correspond to percentiles between 0 and 1 in variable B will be assigned to either 0 or 1, depending upon whether the value is beyond the lower or upper end of the range, respectively.<br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .CAMS_OPI .v0208 .mean .prcp<br>T (Jan 1979) (Dec 2000) RANGE </font> </p> <p style="margin-top: 0"> <font color="#008000">SOURCES .IRI .FD .Seasonal_Forecast .Observations .monthly .prcp </font> </p> <p style="margin-top: 0"> <font color="#008000">T (Jan 1969) (Dec 1998) RANGE </font> </p> <p style="margin-top: 0"> <font color="#008000">T 12 splitstreamgrid </font> </p> <p style="margin-top: 0"> <font color="#008000">[T2]0.0 0.33 0.5 0.67 1. 0.5 replacebypercentile </font> </p> <p style="margin-top: 0"> <font color="#008000">[percentile]CofA=B-bounded</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.CAMS_OPI/.v0208/.mean/.prcp/T/%28Jan%201979%29%28Dec%202000%29RANGE/SOURCES/.IRI/.FD/.Seasonal_Forecast/.Observations/.monthly/.prcp/T/%28Jan%201969%29%28Dec%201998%29RANGE/T/12/splitstreamgrid%5BT2%5D0.0/0.33/0.5/0.67/1./0.5/replacebypercentile%5Bpercentile%5DCofA=B-bounded/">Live Example Link</a> </p> CofA=B-bounded Variable to be converted CofA=B-bounded2 2 false B 1 false A Variable to be converted CofA=B-bounded1 converts a variable A to a variable C using a table B(C) and linear interpolation. Out of range values are pegged to the extreme values. Table Lookup BofA=C-bounded false A BofA=C-bounded1 Variable to be converted 1 4 BofA=C-bounded4 B' <p style="margin-top: 0"> Converted stream using table B(C=A) and linear interpolation </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> Out of range values are pegged to the extreme values </p> false false 2 BofA=C-bounded2 B Variable to be converted BofA=C-bounded Converts a variable A to a variable B using a table B(C=A) and linear interpolation. Out of range values are pegged to the extreme values converts a variable A to a variable B using a table B(C=A) and linear interpolation. Out of range values are pegged to the extreme values. C BofA=C-bounded3 3 false Converts a variable A to a variable B using a table B(C=A) and linear interpolation. Out of range values beyond half a grid step are NaN BofA=C4 B' 4 false New stream using a table B(C=A) and linear interpolation. Out of range values beyond half a grid step are NaN. BofA=C3 3 C false Variable to be converted BofA=C2 false 2 B BofA=C converts a variable A to a variable B using a table B(C=A) and linear interpolation. Out of rangevalues beyond half a grid step are NaN. BofA=C independent variable that is going to be replaced by constant values of <i>S</i> Z false 3 Interpolates onto surfaces A(S) output data: A goes from surfaces of constant Z to surfaces of constant S by linear interpolation 5 S(Z) false 2 input surface information (needs to be monotonic in the independent variable <i>Z</i> so that it can be used as a coordinate) 4 true S value(s) of the new <i>S</i> grid. If omitted, then it uses a list of S values, in which case S needs to have discrete, repeated values. toS input data to be interpolated onto surfaces of <i>S</i> false A(Z) 1 Interpolates onto surfaces false CofA=B2 B Variable to be converted 2 Converts a variable A to a variable B using a table B(C) and linear interpolation. Out of range values beyond half a grid step are NaN CofA=B converts a variable A to a variable C using a table B(C) and linear interpolation. Out of range values beyond half a grid step are NaN. <p style="margin-top: 0"> <br> <b><u>Description</u></b><br><br><b>CofA=B</b> accepts two variables (A and B) as input and converts variable A values to values expressed in terms of variable B based upon a table of corresponding values (B(C)) in variable B and linear interpolation between the points specified in the table. Out-of-range values beyond half a grid step are assigned NaN values.<br><br><b><u>Example</u></b><br><br>In the following example the input variable A is CAMS-OPI gridded monthly precipitation, which depends upon longitude (X), latitude (Y), and months (T) from Jan 1979 to Dec 2000. Input variable B is gridded observed monthly precipitation, which depends upon grids X and Y (which match with X and Y from variable A) and months (T) specified from Jan 1969 to Dec 1998. The purpose of this particular script is to use a specific 30-year set of precipitation values (variable B) to calculate precipitation percentiles by month of year and then express precipitation values from the shorter time series (variable A) in terms of the percentiles from variable B's climatology at each (X,Y) grid point. </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> Here, the time grid for variable B is split into a grid of 12 months per year (T) and a grid of 30 years (T2) from 1969 to 1998. The function &quot;replacebypercentile&quot; is applied to the T2 grid of years in variable B, with the result that the T2 grid is replaced by a new grid named &quot;percentile&quot; that has points at 0.0, 0.33, 0.5, 0.67, and 1, containing the precipitation values that correspond to those percentiles in the 30-year climatology for each month of the year at each (X,Y) grid point. </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> Based upon the correspondence between percentile and precipitation values in variable B function (the table B(C)) &quot;[percentile]CofA=B-bounded&quot; then converts the time series of monthly precipitation values in variable A to percentiles by month of year using linear interpolation between the percentile points specified in the &quot;percentile&quot; grid in variable B. Any precipitation values in variable A that fall outside the range of precipitation values that correspond to percentiles between 0 and 1 in variable B will be assigned missing values (NaN).<br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .CAMS_OPI .v0208 .mean .prcp<br>T (Jan 1979) (Dec 2000) RANGE </font> </p> <p style="margin-top: 0"> <font color="#008000">SOURCES .IRI .FD .Seasonal_Forecast .Observations .monthly .prcp </font> </p> <p style="margin-top: 0"> <font color="#008000">T (Jan 1969) (Dec 1998) RANGE </font> </p> <p style="margin-top: 0"> <font color="#008000">T 12 splitstreamgrid </font> </p> <p style="margin-top: 0"> <font color="#008000">[T2]0.0 0.33 0.5 0.67 1. 0.5 replacebypercentile </font> </p> <p style="margin-top: 0"> <font color="#008000">[percentile]CofA=B</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.CAMS_OPI/.v0208/.mean/.prcp/T/%28Jan%201979%29%28Dec%202000%29RANGE/SOURCES/.IRI/.FD/.Seasonal_Forecast/.Observations/.monthly/.prcp/T/%28Jan%201969%29%28Dec%201998%29RANGE/T/12/splitstreamgrid%5BT2%5D0.0/0.33/0.5/0.67/1./0.5/replacebypercentile%5Bpercentile%5DCofA=B/">Live Example Link</a> </p> CofA=B false A CofA=B1 Variable to be converted 1 CofA=B3 C false 3 Converted variable using table B(C) and linear interpolation Out of range values beyond half a grid step are NaN false CofA=B4 C' 4 CofA=B-bounded returns mean square of A values. If [ grid1 ... ] is given, then the mean AM will meansqover meansqover meansqover1 false A 1 AM 3 meansqover3 false 2 grid meansqover2 false Changes second stream to match truncation of the first maxover Returns the maximum value of a variable over a selected grid(s) Returns the maximum value of a variable over a selected grid(s) true minfrac 3 Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the maximum value to be found. If <i>minfrac</i> is not present, then a missing value is returned. If <i>minfrac</i> is not given, then the maximum value is found regardless of the amount of data present in domain. 4 false maxvar/num maximum value of <i>var</i> within selected domain of <i>grid</i> <p> <i>maxvar/num</i> is no longer dependendent on <i>grid</i>, but is still dependent on any other grids that <i>var</i> depended on (if any) false variable of which maximum value is found 1 var 2 grid false grid(s) over which maximum value is found SOURCES .NOAA .NCEP .CPC .CAMS .station .temperature .temp <br>IWMO (1001) VALUES <br>[T]0.9 maxover <p> OR <p> SOURCES .NOAA .NCEP .CPC .GMSM .w <br>X 112 153 RANGE <br>Y -44 -11 RANGE <br>[X Y]maxover readthredds readthredds url 1 readthredds1 false object 2 false Normally a dataset containing other net references readthredds2 Reads a dataset from a THREDDS server Reads a dataset from THREDDS server NewName false new name of <i>grid</i> 3 SOURCES .NASA .GSFC .TOMS .NIMBUS7 .monthly <br>T (Time) renameGRID ds/var 1 dataset or variable (i.e., data) dependent on grid to be renamed false 4 <i>ds/var</i> dependent on renamed grid false ds_var2 renameGRID Assigns new name to an existing grid (i.e., independent variable) Assigns new name to an existing grid (i.e., independent variable) grid to be renamed 2 grid false variable on which mask is to be applied 1 false var Masks out data values less than a specified threshold SOURCES .NOAA .NCDC .ERSST .version2 .SST <br>28 masklt 2 threshold value on which mask is based false maskval maskedvar false <i>var</i> with data values less than <i>maskval</i> replaced by missing value indicator 3 Flags data values greater than a specified threshold flaggt <br> <u><b>Description</b></u><br><br><b>flaggt</b> replaces values of the input variable &quot;<i>var</i>&quot; with a value of &quot;1&quot; if the input values are greater than the user-specified threshold value &quot;<i>flagval</i>&quot; and replaces values of the input variable &quot;var&quot; with a value of &quot;0&quot; if the input values are less than or equal to the user-specified threshold value &quot;<i>flagval</i>&quot;.<br><br><u><b>Example</b></u><br><br>In the following example <b>flaggt</b> is used to mark those months in which stations in the ANEEL daily precipitation data set have no more than one day of precipitation data missing per month. The &quot;dataflag&quot; command is first applied to the daily precipitation data to flag with a &quot;1&quot; those days with non-missing precipitation data, and to flag with a &quot;0&quot; those days with missing data. &quot;monthlyAverage&quot; is applied to these daily flags of &quot;1&quot;s and &quot;0&quot;s to calculate the fraction of days per month with non-missing data, assigning one value per month. &quot;0.96 <b>flaggt</b>&quot; is applied to these monthly fractional values and replaces them with a value of &quot;1&quot; if at least 96% of days in the month (which is at least all but one day per month, no matter which month) have non-missing precipitation values and a value of &quot;0&quot; if more than one day per month has missing precipitation.<br><br><font color="#008000">SOURCES .ANEEL .prcp_sta .prcp<br>dataflag<br>monthlyAverage<br>0.96 flaggt</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.ANEEL/.prcp_sta/.prcp/dataflag/monthlyAverage/0.96/flaggt/">Live Example Link</a><br> 1 false var variable on which flags are to be applied 3 false binary version of <i>var</i>, where values greater and less than <i>flagval</i> are assigned values of 1 and 0, respectively flaggedvar Flags data values greater than a specified threshold 2 flagval false threshold value used to apply flags 3 false flaggedvar binary version of <i>var</i>, where values greater than and equal to <i>flagval</i> and less than <i>flagval</i> are assigned values of 1 and 0, respectively flagval false threshold value used to apply flags 2 Flags data values greater than or equal to a specified threshold variable on which flags are to be applied false var 1 flagge <br> <u><b>Description</b></u><br><br><b>flagge</b> replaces values of the input variable &quot;<i>var</i>&quot; with a value of &quot;1&quot; if the input values are greater than or equal to the user-specified threshold value &quot;<i>flagval</i>&quot; and replaces values of the input variable &quot;var&quot; with a value of &quot;0&quot; if the input values are less than the user-specified threshold value &quot;<i>flagval</i>&quot;.<br><br><u><b>Example</b></u><br><br>In the following example flagge is used to identify and help count the number of stations in the GHCN precipitation data set that received precipitation greater than or equal to 500 millimeters in December 2006. &quot;<i>500 flagge</i>&quot; replaces precipitation data values of at least 500 mm with a value of &quot;1&quot; and replaces precipitation values below 500 mm with a value of &quot;0&quot;. &quot;<i>[IWMO] sum</i>&quot; sums the flagged data values over the station grid &quot;IWMO&quot; to give a count of the number of stations with at least 500 mm of precipitation.<br><br><font color="#008000">SOURCES .NOAA .NCDC .GHCN .v2beta<br>prcp<br>T (Dec 2006) VALUE<br>500 flagge<br>[IWMO]sum</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.GHCN/.v2beta/prcp/T/%28Dec%202006%29VALUE/500/flagge%5BIWMO%5Dsum/">Live Example Link</a> Flags data values greater than or equal to a specified threshold Flags data values less than a specified threshold var 1 false variable on which flags are to be applied Flags data values less than a specified threshold flagval threshold value used to apply flags false 2 flaggedvar binary version of <i>var</i>, where values less and greater than <i>flagval</i> are assigned values of 1 and 0, respectively false 3 <br> <u><b>Description</b></u><br><br><b>flaglt</b> replaces values of the input variable &quot;<i>var</i>&quot; with a value of &quot;1&quot; if the input values are less than the user-specified threshold value &quot;<i>flagval</i>&quot; and replaces values of the input variable &quot;var&quot; with a value of &quot;0&quot; if the input values are greater than or equal to the user-specified threshold value &quot;<i>flagval</i>&quot;.<br><br><u><b>Example</b></u><br><br>In the following example &quot;12 <b>flaglt</b>&quot; is applied to monthly climatological precipitation values from the UEA TS2.1 data set to denote very dry locations that receive less than 12 mm of precipitation per month, on average.<br><br><font color="#008000">SOURCES .UEA .CRU .TS2p1 .climatology .c7100 .prcp <br>12 flaglt</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.UEA/.CRU/.TS2p1/.climatology/.c7100/.prcp/12/flaglt/">Live Example Link</a><br> flaglt variable on which flags are to be applied 1 false var Flags data values less than and equal to a specified threshold flagle flagval 2 false threshold value used to apply flags Flags data values less than and equal to a specified threshold <br> <u><b>Description</b></u><br><br><b>flagle</b> replaces values of the input variable &quot;<i>var</i>&quot; with a value of &quot;1&quot; if the input values are less than or equal to the user-specified threshold value &quot;<i>flagval</i>&quot; and replaces values of the input variable &quot;var&quot; with a value of &quot;0&quot; if the input values are greater than the user-specified threshold value &quot;<i>flagval</i>&quot;.<br><br><u><b>Example</b></u><br><br>In the following example, &quot;28 flagle&quot; replaces monthly sea surface temperature values less than or equal to 28&#176;C with a &quot;1&quot; and SST values greater than 28&#176;C with a &quot;0&quot;.<br><br><font color="#008000">SOURCES .NOAA .NCDC .ERSST .version2 .SST<br>28 flagle</font><br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.ERSST/.version2/.SST/28/flagle/">Live Example Link</a><br> flaggedvar binary version of <i>var</i>, where values less than or equal to <i>flagval</i> and greater than <i>flagval</i> are assigned values of 1 and 0, respectively false 3 Masks out data values less than a specified threshold masklt Ends a Lorenz 4-cycle scheme integration loop 1 false L4cycle:endLoop1 dh L4cycle:endLoop2 false 2 h L4cycle:endLoop L4cycle:endLoop ends a Lorenz 4-cycle scheme integration loop Computes the (transposed) generalized inverse with the ivar1 grids as the first dimentsion of the matrix and the ivar2 grids as the second dimension of the matrix false First dimension of the matrix 2 ivar1 3 false Second dimension of the matrix ivar2 inverse Generalized inverse false 4 var false 1 Variable from which inverse will be computed ginverse Fourier_transform computes fouriertransform Fourier Transform 1 or -1 false 3 fourier transform3 sign var' fourier transform4 4 false stream after fouier transform computed Variable to be transformed var 1 false fourier transform1 Domain to be transformed to frequency or wavelength ivar fourier transform2 false 2 Computes fouriertransform false 7 normalized frequency distribution of <i>var</i> within range and intervals specified by <i>lower</i>, <i>upper</i>, and <i>step</i> dist upper upper bound of range over which distribution is to be found 4 false false 2 DATA Returns the normalized frequency distribution of a set of data for a specified range and step interval. 3 lower false lower bound of range over which distribution is to be found Returns the normalized frequency distribution of a set of data for a specified range and step interval. normalizeddistrib1D width of intervals (in units of var) used to find distribution 5 step false SOURCES .NOAA .NCEP .CPC .GMSM .w <br>Y (20) VALUE <br>X (50) VALUE <br>T (Jul 1948-2003) VALUES <br>DATA 0 50 2 RANGESTEP <br>normalizeddistrib1D variable (i.e., data) of which distribution is to be found false 1 var 6 false RANGESTEP Variable converted to random data uniformly distributed onto gaussian data false 2 var' 1 Variable to be converted var false gaussianinv Converts random data uniformly distributed on to gaussian data sqrt Calculates the square root of a number or variable Calculates the square root of a number or variable Result square root of <i>var/num</i>. <p> Result will be of the same type as <i>var/num</i> (i.e., either a variable or constant). 2 1 variable or constant of which the square root is to be taken var/num SOURCES .UEA .CRU .New .CRU05 .monthly .prcp <br> sqrt converts daily data to dekadal by averaging <p style="margin-top: 0"> <br> <u><b>Description</b></u><br><br>A dekad is a unit of time with a peculiar definition. There are three dekads in a calendar month. The first ten days of a month constitute the first dekad of the month. The second ten days constitute the second dekad of the month, and the remaining days (8 to 11 days, depending upon the month) constitute the third dekad.<br><br><b>dekadalAverage</b> calculates dekadal average values from an input variable containing daily data.<br><br><u><b>Example</b></u><br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .FEWS .SAsia .RFEv2 .DAILY .est_prcp<br>T (1 Jan 2007) (30 Apr 2007) RANGE<br>dekadalAverage</font><br><br>This example selects daily precipitation estimates from the CPC/FEWS South Asia RFE product for the period 1 January to 30 April 2007 and calculates the dekadal averages of the daily values such that there are average precipitation values for 1-10 January, 11-20 January, 21-31 January, 1-10 February 2007, etc. The underlying time grid remains a daily grid, but the average values are assigned to the dekadal intervals.<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.FEWS/.SAsia/.RFEv2/.DAILY/.est_prcp/T/%281%20Jan%202007%29%2830%20Apr%202007%29RANGE/dekadalAverage/">Live Example Link</a><br> </p> dekadalAverage daily data to be averaged, with units of units false dailyvar 1 Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within each dekad in order for each average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the average is calculated regardless of the amount of data present. 2 minfrac true false data averaged by dekad dekadalvar 3 converts daily data to dekadal by averaging Masks out all values of a variable not included in the indicated range. Commonly used to select stations within a particular lat/lon range (see Live Example). 3 range_max upper threshold of range variable on which mask will be applied 1 variable restricted_var <i>variable</i> with all values outside of range specified by <i>range_min</i> and <i>range_max</i> masked out 4 masknotrange range_min 2 lower threshold of range Masks out all values of a variable not included in the indicated range. Commonly used to select stations within a particular lat/lon range (see Live Example). <u><b>Description</b></u><br><font color="#008000"><b>masknotrange</b></font><font color="#000000"> masks out all values of a variable not included in the indicated range. Commonly used to select stations within a particular lat/lon range.</font><font color="#008000"><br></font><u><b>Example</b></u><br><font color="#008000">lat -20 20 masknotrange</font><br><font color="#800000"><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.GDCN/lon/290/340/masknotrange/SELECT/lat/-10/10/masknotrange/SELECT/">Live Example Link</a></font> SOURCES .NOAA .NCEP-NCAR .CDAS-1 .MONTHLY .Intrinsic .PressureLevel <br>P <br>grid: <br>/name (Pressure) def <br>/units (hPa) def <br>values: 1000 925 850 700 600 500 400 300 250 200 150 100 70 50 30 20 10 :values <br>:grid <br>replaceGRID grid1 false grid (i.e., independent variable) to be replaced 2 replaceGRID data 1 false dataset or variable dependent on grid to be replaced 4 <i>data</i> with <i>grid2</i> in place of <i>grid1</i> <p> Note that no changes are made to the data itself. false data_new Replaces a grid (i.e., independent variable) 3 grid2 grid (i.e., independent variable) to replace <i>grid1</i>. Can be an existing or newly-defined grid. false Replaces a grid (i.e., independent variable) rotatedata3 Y 3 false 9 Rotated variable false rotatedata9 D' 7 false rotatedata7 ox rotatedata6 6 alpha false rotatedata rotatedata4 false 4 X' rotatedata 5 rotatedata5 false Y' Rotates a variable onto a new grid (X'Y') rotates a variable onto a new grid (X'Y') 1 Variable to be rotated rotatedata1 false D 8 false rotatedata8 oy false rotatedata2 2 X Inverts onto surfaces invertontogrid Inverts onto surfaces independent variable that is going to be replaced by S Z false 2 S(Z) false 1 input surface information (needs to be monotonic in the independent variable <i>Z</i> so that it can be used as a coordinate) false value(s) of the new <i>S</i> grid S 3 Z(S) 4 output data: <i>Z</i> values for surfaces of constant <i>S</i> normalize normalize Divides var1 by var2. Points less than minimum in var2 become NaN in the output divides var1 by var2. Points less than minimum in var2 become NaN in the output false var1 1 normalize1 Variable to be divided by var2 false normalize2 2 var2 Variable dividing var1 normalize3 Minimum value allowed before being replaced with NaN minimum 3 false Var1 divided by var2 Points less than minimum in var2 replaced with NaN var' normalize4 4 false <br> <b><u>Description</u></b><br><br><b>correlate</b> calculates the Pearson product moment correlation for the two latest items on the stack over the indicated grid. For the correlation to be computed, the gridding of the two items on the stack must match.<br><br><b><u>Example</u></b><br><br><font color="#008000">SOURCES .NOAA .NCEP .CPC .GMSM .w<br>T (Jan 1969) (Dec 1998) RANGE<br>X (-8) (20) RANGE<br>Y (8) (20) RANGE<br>SOURCES .DEKLIM .VASClimO .PrcpClim .Resolution-0p5x0p5 .prcp<br>T (Jan 1969) (Dec 1998) RANGE<br>X (-8) (20) RANGE<br>Y (8) (20) RANGE<br>[T]correlate</font><br><br>In this example, GMSM monthly soil moisture values are correlated over the time grid with monthly precipitation values from the VASClim0 data set for a region of West Africa over the period January 1969 to December 1998. The gridding of the two data sets matches in both space (0.5 deg. lat/lon resolution) and time. The result is a single map of correlation coefficients over the defined region of West Africa.<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.GMSM/.w/T/%28Jan%201969%29%28Dec%201998%29RANGE/X/%28-8%29%2820%29RANGE/Y/%288%29%2820%29RANGE/SOURCES/.DEKLIM/.VASClimO/.PrcpClim/.Resolution-0p5x0p5/.prcp/T/%28Jan%201969%29%28Dec%201998%29RANGE/X/%28-8%29%2820%29RANGE/Y/%288%29%2820%29RANGE%5BT%5Dcorrelate/">Live Example Link</a><br> <p> OR </p> <p> <font color="#008000">SOURCES .NOAA .NCEP .CPC .CAMS_OPI .v0208 .anomaly .prcp<br>T (Jan 1980) (Dec 2003) RANGE<br>X (-150) (-80) RANGE<br>Y (-10) (10) RANGE<br>SOURCES .NOAA .NCEP .EMC .CMB .GLOBAL .Reyn_SmithOIv2 .monthly .sst<br>T (Jan 1980) (Dec 2003) RANGE<br>X (-150) (-80) RANGE<br>Y (-10) (10) RANGE<br>[X Y]regridAverage<br>[X Y]correlate</font><br><br>In this example, CAMS_OPI monthly precipitation data are correlated in space with Reynolds and Smith OI Version 2 monthly sea surface temperatures over the equatorial Pacific Ocean. The result is a monthly time series of values from January 1980 to December 2003. The [X Y]regridAverage command was used to spatially regrid the SST data set (at 1.0 deg. lat/lon resolution) to match the gridding of the precipitation data set (at 2.5 deg. lat/lon resolution) before taking the correlation.<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCEP/.CPC/.CAMS_OPI/.v0208/.anomaly/.prcp/T/%28Jan%201980%29%28Dec%202003%29RANGE/X/%28-150%29%28-80%29RANGE/Y/%28-10%29%2810%29RANGE/SOURCES/.NOAA/.NCEP/.EMC/.CMB/.GLOBAL/.Reyn_SmithOIv2/.monthly/.sst/T/%28Jan%201980%29%28Dec%202003%29RANGE/X/%28-150%29%28-80%29RANGE/Y/%28-10%29%2810%29RANGE%5BX/Y%5DregridAverage%5BX/Y%5Dcorrelate/">Live Example Link</a><br><br><br> </p> variable to be correlated with <i>var2</i> var1 1 false correlate false coefficient Pearson-Product Moment Correlation coefficient of <i>var1</i> and <i>var2</i> over <i>grids</i>. <p> <i>coefficient</i> is not dependent on <i>grids</i>, but is dependent on any other grids that <i>var1</i> or <i>var2</i> depended on (if any). 5 Calculates the Pearson Product-Moment Correlation coefficient of two variables over specified grids (i.e., independent variables) 2 false var2 variable to be correlated with <i>var1</i> <p> Note that <i>var1</i> and <i>var2</i> should have similarly-defined <i>grids</i>. Regridding one variable to match the other may be necessary (see example below). false grid(s) (i.e., independent variables) over which correlation coefficient is to be calculated grids 3 Calculates the Pearson Product-Moment Correlation coefficient of two variables over specified grids (i.e., independent variables) 4 minfrac Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the correlation to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the correlation is calculated regardless of the amount of data present in the domain. true Sample by Variable Samples variable1 along grid using variable2=grid false samplealong2 2 var2 Variable to be used samples variable1 along grid using variable2=grid false 1 sample-along1 var1 Variable to be sampled false Variable after being sampled along grid 4 sample-along4 var1' sample-along sample-along 3 sample-along3 false Independent variable=variable2 ivar Bins data into bins from lo to high by step, returning a mask with an additional grid (the bins) that is either NaN or 1 bins data into bins from lo to high by step, returning a mask with an additional grid (the bins) that is either NaN or 1 maskbybin4 Mask with additional grid (the bins) that is either NaN or 1 3 var' false bin edges 2 maskbybin2 false Variable to be sorted var false maskbybin1 1 maskbybin maskbybin Shifts a grid (i.e., independent variable) by a specified number of grid points shiftGRID 3 false number of grid points the grid is to be shifted. <i>num</i>>0 (<i>num</i><0) shifts grid forwards (backwards) num false data dependent on the grid (i.e., independent variable) to be shifted 1 ds/var same as <i>ds/var</i> with <i>grid</i> relabeled according to a shift of <i>num</i> grid points ds/varshift false 4 SOURCES .IRI .FD .ECHAM4p5 .Forecast .psst .ensemble12 .MONTHLY .PressureLevel-SF .phi <br>M -12 shiftGRID <p>OR<p> SOURCES .IRI .Analyses .SPI .SPI-CMAP0407v1_3-Month <br>T /pointwidth 1 def <br>pop <br>T 1 shiftGRID Shifts a grid (i.e., independent variable) by a specified number of grid points grid (i.e., independent variable) to be shifted 2 grid false 2 variable to be rank correlated with <i>var1</i> <p> Note that <i>var1</i> and <i>var2</i> should have similarly-defined <i>grids</i>. Regridding one variable to match the other may be necessary (see example below). false var2 SOURCES .NOAA .NCEP .CPC .GSOD .MONTHLY .DATA .mean .pan .evaportransporation <br>SOURCES .NOAA .NCEP .CPC .GSOD .MONTHLY .DATA .maximum .temp <br>IWMO (105130) VALUE <br>[T]rankcorrelate <p> OR <p> SOURCES .NOAA .NCEP .CPC .CAMS_OPI .v0208 .anomaly .prcp <br>T (Jan 1980) (Dec 2003) RANGE <br>X (-150) (-80) RANGE <br>Y (-10) (10) RANGE <br>SOURCES .NOAA .NCEP .EMC .CMB .GLOBAL .Reyn_SmithOIv2 .monthly .sst <br>T (Jan 1980) (Dec 2003) RANGE <br>X (-150) (-80) RANGE <br>Y (-10) (10) RANGE <br>[X Y]regridAverage <br>[X Y]rankcorrelate 1 variable to be rank correlated with <i>var2</i> var1 false true minfrac 4 Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the rank correlation to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the rank correlation is calculated regardless of the amount of data present in the domain. false grids grid(s) (i.e., independent variables) over which rank correlation coefficient is to be calculated 3 Calculates the Spearman rank correlation coefficient of two variables over specified grids (i.e., independent variables) rankcorrelate Calculates the Spearman rank correlation coefficient of two variables over specified grids (i.e., independent variables) coefficient Spearman rank correlation coefficient of <i>var1</i> and <i>var2</i> over <i>grids</i>. <p> <i>coefficient</i> is not dependent on <i>grids</i>, but is dependent on any other grids that <i>var1</i> or <i>var2</i> depended on (if any). 5 false 1 dataset to be sorted dataset false false grid independent variables to be replaced by rank 3 false 2 variable 4 sorteddataset sorted dataset where <i>grid </i>is replaced by<i> rank. </i>Rank value 1 corresponds to the maxium value of variable. false sort_by sorts dataset variables by replacing grid with rank, rank being calculated from variable. false Variable to be sampled along 1 var range_low 3 Lower threshold of range (center of closest grid box) false 2 false ivar Independent variable to be applied to variable 5 false var' Variable sampled along one coordinate false range_high Upper threshold of range (center of closest grid box) 4 RANGESPAN Samples variable along one coordinate. Works by creating a child object of the variable X 2 false isolines2 isolines3 3 Y false Variable from which data will be computed from isolines1 false var 1 computes isolines from a gridded dataset. Should be changed to output a geometry. isolines Computes isolines from a gridded dataset. Should be changed to output a geometry isolines7 false 7 GRDS isolines6 false 6 YS XS false isolines5 5 isolines levels isolines4 false 4 <p style="margin-top: 0"> computes complementary error function from variable </p> erfc C false 2 A 1 false selectbyNaN Replaces missing_values in variable2 with variable1. Documentation carried primarily from variable1, however, making it a bit different that a reversed replaceNaN replaces missing_values in variable2 with variable1. Documentation carried primarily from variable1, however, making it a bit different that a reversed replaceNaN. selectbyNaN false var1' Variable with missing_values replaced selectbyNaN3 3 selectbyNaN1 var1 false Variable from which missing values will be replaced 1 2 false var2 selectbyNaN2 Variable from which missing values will be replaced with true keepgrids keeps independent variables averaged over in output as single values grids 4 variable to be averaged 1 var false Calculates the average false avgvar/num 5 <i>var</i> averaged over <i>grid</i>. <p><i>avgvar</i> is no longer dependent on <i>grid</i>, but is still dependent on any other grids that <i>var</i> depended on (if any) Calculates the average Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the average to be calculated. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the average is calculated regardless of the amount of data present in domain. true minfrac 3 average grid(s) over which the average is to be calculated 2 false grid average <br> <b><u>Description</u></b><br><br><b>average</b> calculates a simple average, over the grid(s) included in the brackets, of the values in the latest variable on the stack. The optional <i>minfrac</i> argument is a number between 0. and 1.0 that indicates what fraction of data must be non-missing over the averaging grid in order for a non-missing result to be produced. If <i>minfrac</i> is not specified, its default value is 0., and an average is produced using all available data over the averaging grid.<br><br><b><u>Example</u></b><br><br><font color="#008000">SOURCES .NOAA .NCDC .ERSST .version2 .SST<br>X (170W) (120W) RANGE<br>Y (5S) (5N) RANGE<br>[X Y]average</font><br><br>This example averages over both longitude (X) and latitude (Y) to produce an average sea surface temperature value (for each time step) over the equatorial Pacific Ocean.<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.ERSST/.version2/.SST/X/%28170W%29%28120W%29RANGE/Y/%285S%29%285N%29RANGE%5BX/Y%5Daverage/">Live Example Link</a> <p> OR </p> <p> <font color="#008000">SOURCES .NOAA .NCDC .GHCN .v2beta .prcp<br>IWMO 60354001 VALUE<br>T (Jan 1960) (Dec 1979) RANGE<br>[T]0.5 average</font><br><br>This example produces the average monthly precipitation value for Algiers, Algeria, over the period January 1960 to December 1979, if at least half the monthly precipitation values over that time period are not missing.<br><br><a href="http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.GHCN/.v2beta/.prcp/IWMO/60354001/VALUE/T/%28Jan%201960%29%28Dec%201979%29RANGE%5BT%5D0.5/average/">Live Example Link</a> </p> grid set describing domain to be detrended over grids 2 false detrends with a best-fit-line 3 minfrac true Minimum fraction of data that must be present (i.e., fraction not indicated as missing) within the selected domain in order for the data to be used in the best-fit. If minfrac is not present, then a missing value is returned. If minfrac is not given, then the bestfit is calculated regardless of the amount of data present in the domain. <p style="margin-top: 0"> While detrending data can be statistically important, it is important to remember that you could be removing the primary signal. For example, consider </p> <p style="margin-top: 0"> <img src="http://iridl.ldeo.columbia.edu/SOURCES/.KEELING/.MAUNA_LOA/.co2/T+fig-+line+-fig//plotborder+72+psdef//plotaxislength+432+psdef+.gif"> </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> where the point is that there is a long-term trend in the CO2 levels. One can, however, easily detrend the line with </p> <pre>SOURCES .KEELING .MAUNA_LOA .co2 [T]detrend-bfl</pre> <p style="margin-top: 0"> which would result in </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> <img src="http://iridl.ldeo.columbia.edu/expert/SOURCES/.KEELING/.MAUNA_LOA/.co2%5BT%5Ddetrend-bfl/T+fig-+line+-fig//plotborder+72+psdef//plotaxislength+432+psdef+.gif"> </p> <p style="margin-top: 0"> </p> <p style="margin-top: 0"> Note that if we were dealing with a multi-dimensional datasets (such as X Y T representing longitude, latitute, and time), then applying [T] detrend-bfl results in a separate line being calculated at each spatial point. </p> detrend-bfl variable to be detrended false var 1 detrended version of the variable false 4 detrended-var Detrending detrend-bfl ends an explicit integration loop Explicit:endLoop Ends an explicit integration loop Explicit:endLoop 1 dh false Explicit:endLoop1 false Explicit:endLoop2 h 2 computes inverse error function from variable false A 1 C false 2 erfinv Masks out all values of a variable included in the indicated range. range_min lower threshold of range 2 restricted_var 4 <i>variable</i> with all values inside range specified by <i>range_min</i> and <i>range_max</i> masked out 3 upper threshold of range range_max Masks out all values of a variable included in the indicated range. variable variable on which mask will be applied 1 maskrange lat -20 20 maskrange Masks out data values greater than a specified threshold 1 var variable on which mask is to be applied false maskgt false maskvar 3 <i>var</i> with data values greater than <i>maskval</i> replaced by missing value indicator Masks out data values greater than a specified threshold maskval 2 false threshold value on which mask is based SOURCES .NOAA .NCEP .CPC .GMSM .w <br>100 maskgt