# # File: # panel1.py # # Synopsis: # Illustrates how to draw several visualizations on a single frame. # # Category: # Paneling # Contours over maps # Label bar # # Author: # Fred Clare (based on a code of Mary Haley) # # Date of initial publication: # September, 2004 # # Description: # Illustrates how to put multiple plots on a page and how # to have each plot have its own label bar or have a common # label bar for all plots on the page. # # Effects illustrated: # o How to put multiple plots on a page. # o Drawing contours over maps. # o Some special paneling resources. # o Label bars on individual plots and a common label bar for all plots. # o How to put a label on each individual plot on a multi-plot frame. # # Output: # This examples produces three visualizations: # 1.) Four contour plots on a single frame with individual label bars. # 2.) Same as 1.), but with additional white space around the plots. # 3.) Six plots on a page with labels applied and a common label bar. # # Notes: # # # Import numpy. # from __future__ import print_function import numpy, os # # Import Ngl support functions. # import Ngl # # Import Nio. # import Nio # # Open a netCDF file containing storm data. # dirc = Ngl.pynglpath("data") tfile = Nio.open_file(os.path.join(dirc,"cdf","Tstorm.cdf")) # # Access the temperature arrays for the first 6 time steps. # nplots = 6 temp = tfile.variables["t"] # # Save lat and lon to numpy arrays. # lat = tfile.variables["lat"][:] lon = tfile.variables["lon"][:] # # Define a color map. # cmap = numpy.array([[1.00,.000,.000],\ [.950,.010,.000],[.870,.050,.000],[.800,.090,.000],\ [.700,.090,.000],[.700,.120,.000],[.700,.180,.000],\ [.700,.260,.000],[.700,.285,.000],[.680,.330,.000],\ [.570,.420,.000],[.560,.530,.000],[.550,.550,.000],\ [.130,.570,.000],[.060,.680,.000],[.000,.690,.000],\ [.000,.700,.100],[.000,.600,.300],[.000,.500,.500],\ [.000,.400,.700],[.000,.300,.700],[.000,.200,.700],\ [.000,.100,.700],[.000,.000,.700],[.100,.100,.700],\ [.200,.200,.700],[.300,.300,.700],[.420,.400,.700],\ [.560,.500,.700],[.610,.600,.700],[.700,.700,.700]], 'f') #--- Indicate where to send graphics rlist = Ngl.Resources() wks_type = "png" wks = Ngl.open_wks(wks_type,"panel1",rlist) # # Turn off draw for the individual plots, since we are going to # panel them later. # resources = Ngl.Resources() resources.nglDraw = False resources.nglFrame = False # # Loop through the timesteps and create each plot, titling each # one according to which timestep it is. # plot = [] resources.cnFillOn = True # Turn on contour fill. resources.cnFillPalette = "MPL_viridis" resources.cnLineLabelsOn = False # Turn off contour labels # # Set some font heights to make them slightly bigger than the default. # Turn off nglScale, because this resource wants to set the axes font # heights for you. # resources.nglScale = False resources.tiMainFontHeightF = 0.037 resources.lbLabelFontHeightF = 0.032 resources.tmXBLabelFontHeightF = 0.030 resources.tmYLLabelFontHeightF = 0.030 for i in range(0,nplots): resources.tiMainString = "Temperature at time = {}".format(i) plot.append(Ngl.contour(wks,temp[i,:,:],resources)) Ngl.panel(wks,plot[0:4],[2,2]) # Draw 2 rows/2 columns of plots. # # Now add some extra white space around each plot. # panelres = Ngl.Resources() panelres.nglPanelYWhiteSpacePercent = 5. panelres.nglPanelXWhiteSpacePercent = 5. Ngl.panel(wks,plot[0:4],[2,2],panelres) # Draw 2 rows/2 columns of plots. # # This section will set resources for drawing contour plots over a map. # del resources.tiMainString # Don't set a main title. resources.cnFillPalette = cmap # Set color map using RGB values resources.sfXArray = lon # Portion of map on which to overlay resources.sfYArray = lat # contour plot. resources.cnLineLabelsOn = False # Turn off contour line labels. resources.cnLinesOn = False # Turn off contour lines. resources.cnFillOn = True # Turn on contour fill. resources.cnLevelSelectionMode = "ManualLevels" # Select contour levels. resources.cnMinLevelValF = 245. resources.cnMaxLevelValF = 302.5 resources.cnLevelSpacingF = 2.5 resources.tmXBLabelFontHeightF = 0.020 resources.tmYLLabelFontHeightF = 0.020 resources.mpLimitMode = "LatLon" # Limit portion of map that is viewed. resources.mpMinLatF = float(min(lat)) resources.mpMaxLatF = float(max(lat)) resources.mpMinLonF = float(min(lon)) resources.mpMaxLonF = float(max(lon)) resources.pmLabelBarDisplayMode = "Never" # Turn off labelbar, since we # will use a global labelbar # in the panel. resources.mpPerimOn = True # Turn on map perimeter. resources.mpGridAndLimbOn = False # Turn off map grid. plot = [] for i in range(0,nplots): plot.append(Ngl.contour_map(wks,temp[i,:,:],resources)) # # Draw two titles at the top. Draw these before the panel stuff, # so that the maximization works properly. # textres = Ngl.Resources() textres.txFontHeightF = 0.025 # Size of title. Ngl.text_ndc(wks,"~F22~Temperature (K) at every six hours",0.5,.97,textres) textres.txFontHeightF = 0.02 # Make second title slightly smaller. Ngl.text_ndc(wks,"~F22~January 1996",0.5,.935,textres) # # Set some resources for the paneled plots. # del panelres panelres = Ngl.Resources() # # Set up some labelbar resources. Set nglPanelLabelBar to True to # indicate you want to draw a common labelbar at the bottom of the # plots. # panelres.nglPanelLabelBar = True # Turn on panel labelbar panelres.nglPanelLabelBarLabelFontHeightF = 0.015 # Labelbar font height panelres.nglPanelLabelBarHeightF = 0.1750 # Height of labelbar panelres.nglPanelLabelBarWidthF = 0.700 # Width of labelbar panelres.lbLabelFont = "helvetica-bold" # Labelbar font panelres.nglPanelTop = 0.935 panelres.nglPanelFigureStrings = ["A","B","C","D","E","F"] panelres.nglPanelFigureStringsJust = "BottomRight" # # You can have PyNGL selection the best paper orientation for # the shape of plots you are drawing. This resource is for PDF or # PS output only. # if(wks_type == "ps" or wks_type == "pdf"): panelres.nglPaperOrientation = "Auto" # # Draw 3 rows and 2 columns of plots. # Ngl.panel(wks,plot[0:nplots],[3,2],panelres) Ngl.end()