RasterHistograms

Package workings

For a single Raster (i.e. one variable)

julia> dummy_data = repeat(1:10; outer = (1, 10));

julia> rs = Raster(dummy_data, (X(1:10), Ti(1:10)); name = :dummy_variable);

julia> rs_hist = RasterLayerHistogram(rs)
RasterLayerHistogram for the variable dummy_variable
┣━━ Layer dimensions: (:X, :Ti) 
┣━━━━━━━━ Layer size: (10, 10)
┗━━━━━━━━━ Histogram: 1-dimensional

julia> rs_hist.histogram
StatsBase.Histogram{Int64, 1, Tuple{StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}}}
edges:
  0.0:2.0:12.0
weights: [10, 20, 20, 20, 20, 10]
closed: left
isdensity: false

a one dimensional Histogram that has been fit to the dumy_variable data is returned in the rs_hist.histogram field as well as some information about the data the Histogram was fit to. If a RasterStack or RasterSeries with multiple layers is passed in the default behaviour is to fit an N-dimensional Histogram where N is the number of layers (i.e. the number of variables).

julia> vars = (v1 = randn(10, 10), v2 = randn(10, 10), v3 = randn(10, 10));

julia> stack = RasterStack(vars, (X(1:10), Y(1:10)))
RasterStack with dimensions: 
  X Sampled{Int64} 1:10 ForwardOrdered Regular Points,
  Y Sampled{Int64} 1:10 ForwardOrdered Regular Points
and 3 layers:
  :v1 Float64 dims: X, Y (10×10)
  :v2 Float64 dims: X, Y (10×10)
  :v3 Float64 dims: X, Y (10×10)


julia> RasterStackHistogram(stack)
RasterStackHistogram for the variables (:v1, :v2, :v3)
┣━━ Stack dimensions: (:X, :Y)
┣━━ Stack layer size: (10, 10)
┗━━━━━━━━━ Histogram: 3-dimensional

Info

The order of the variables for the Histogram is the order of the layers in the RasterStack or RasterSeries. This can be important for plotting when variables are desired to be on specific axes. In the example above v1 would be on the x-axis, v2 the y-axis and v3 the z-axis. To change which axes the variables correspond to, the order of the layers in the RasterStack would need to be altered (or you could plot from the histogram.weights matrix).

Plotting

Both Makie.jl and Plots.jl have functions in the module to extract the Histogram object from the AbstractRasterHistogram for plotting. To plot in either package one can just call

julia> using #Plotting package e.g. CairoMakie.jl or PLots.jl 

julia> plot(::AbstractRasterHistogram)

and an N-dimensional Histogram will be plotted where N is the dimension of the ::AbstractRasterHistogram. Makie.jl is used in the exmaple.

For a full list of the functions in this module see the function index or look at the example to see the module in action.