Miscellaneous functions

EQV(T::Unitful.Temperature) -> Unitful.Pressure

Calculates equilibrium vapor pressure from temperature T, which can by any Unitful temperature quantity, as it is internally converted to Kelvin. Returns the value in units of Pa.

For details see Murphy & Koop (2005): doi:10.1256/qj.04.94.

FWHM_to_sigma(FWHM::Real) -> Any

Calculates the standard deviation of a Gaussian for a given full width at half the maximum value.

H2O_VMR_to_relative_humidity(
    H2O_VMR::Real,
    p::Unitful.Pressure,
    T::Unitful.Temperature
) -> Number

Calculates relative humidity given a H₂O volume mixing ratio H2O_VMR, pressure p, and temperature T. Converts the result to Unitful.NoUnits, thus returning a number.

julia> H2O_VMR_to_relative_humidity(0.025, 1013e2u"Pa", 300u"K") ≈ 0.71604995533615401
true

H2O_VMR_to_specific_humidity(h2o::Number) -> Any

Calculates specific humidity from H2O VMR

W_to_ph(
    L_in::Unitful.Quantity,
    ν::Unitful.Wavenumber
) -> Unitful quantity with units ph s⁻¹ m⁻² sr⁻¹ (cm⁻¹)⁻¹

Converts a spetral radiance in units of power per area, per steradian per wavelength into spectral radiance in units of ph s⁻¹ m⁻² sr⁻¹ (cm⁻¹)⁻¹. Users can then cast this into a quantity of other wavelength units. The function interface allows L_in to be any Unitful quantity, however the function checks whether L_in is dimensionally compatible with spectral radiance type units of power per wavenumber.

Example

W_to_ph(
    L_in::Unitful.Quantity,
    λ::Unitful.Length
) -> Unitful quantity with units ph μm^-1 m^-2 s^-1 sr^-1

Converts a spetral radiance L_in in units of power per area, per steradian per wavelength into spectral radiance in units of ph μm^-1 m^-2 s^-1 sr^-1. Users can then cast this into a quantity of other wavelength units. The function interface allows L_in to be any Unitful quantity, however the function checks whether L_in is dimensionally compatible with spectral radiance type units of power per wavelength.

Example

avx_add_along_columns!(
    y::AbstractVector,
    x::AbstractMatrix
)

Adds the along-column sum of matrix x to the vector y. This function makes use of the @turbo macro of the LoopVectorization package for high performance. The contents of y are zeroed out prior to the computation. Equivalent to y[:] .+= sum(x, dims=2), but roughly 3x faster.

avx_sum_along_columns_between!(
    y::AbstractVector,
    x::AbstractMatrix,
    idx1::Integer,
    idx2::Integer
)

Calculates the along-column sum of matrix x between indices idx1 and idx2, to store it in the vector y. It is assumed that idx1 < idx2. This function makes use of the @turbo macro of the LoopVectorization package for high performance. The contents of y are zeroed out prior to the computation. Equivalent to y[:] .= sum(x[:,idx1:idx2], dims=2), but roughly 2x faster.

Example

Below example shows a matrix with two rows and three columns. avx_sum_along_columns_between! computes the sums for each of the two rows, between column indices 2 and 3, meaning 2+3 and 5+6, and stores the results in the two-element vector y.

x = [1 2 3; 4 5 6]
y = zeros(Int, 2)
avx_sum_along_columns_between!(y, x, 2, 3)
y

check_for_not_finite(x::AbstractArray) -> Bool

For an array x, returns true if any element is not finite (as per isfinite), or false if all elements are finite.

findany(x::AbstractVector, a) -> Bool

Returns true if a is found within x, does not allocate.

Examples

julia> findany(["a", "b", "c", "d"], "b")
true

julia> findany([1,2,3,4,5], 10)
false

findanytype(x::AbstractVector, T) -> Bool

Returns true if any element within x is of type T, does not allocate.

ingest!(obj, field::Symbol, val::AbstractArray)

Sets a RetrievalToolbox type field and applies the correct unit conversion. This is the specialized function to deal with array-type values. This should be mostlya allocation-free.

Details

ingest!(obj, field::Symbol, val::Number)

Sets a RetrievalToolbox type field and applies the correct unit conversion

Details

levels_to_layers!(
    layers::AbstractVector,
    levels::AbstractVector;
    logspace
)

For a given level vector, calculate the mid-layer values and store into layers. This function assumes that the layer-value is evaluated at the center point between the two adjacent levels.

levels_to_layers(levels::AbstractVector; logspace) -> Any

For a given level object, create a new vector that contains the mid-layer vector. This function assumes that the layer-value is evaluated at the center point between the two adjacent levels.

maxmin(x::AbstractArray) -> Any

For an array x, returns the difference between the largest and smallest element maximum(x) - minimum(x).

a * b

specific_humidity_to_H2O_VMR(sh::Number) -> Any

Calculates H2O VMR from specific humidity