Estimates CO2 efflux (e.g., soil respiration (Rs), [μmol m-2 s-1]) as a function of temperature (e.g., soil temperature (Ts), [°C]) and soil moisture (θ, [m3 m-3]).
Install with the Julia package manager Pkg, just like any other registered Julia package:
julia> ]
pkg> add DAMMmodelor
julia> using Pkg; Pkg.add("DAMMmodel")This package models respiration (CO2 efflux, e.g., soil respiration (Rs)) as a function of soil temperature (Ts) and soil moisture (θ), using the Dual Arrhenius and Michaelis-Menten kinetics model (2012).
The package contains six functions: DAMMviz, DAMMfdata, DAMM, DAMMfit, DAMMmat, and DAMMplot.
DAMMviz()
Interactive plot of the DAMM model
julia> DAMMviz()
DAMMfdata(n)
Generates a DataFrame of n fake data Tₛ, θ and Rₛ
julia> DAMMfdata(5)
5×3 DataFrame
Row │ Tₛ θ Rₛ
│ Float64 Float64 Float64
─────┼───────────────────────────
1 │ 10.8 0.3 2.04327
2 │ 31.5 0.1 7.8925
3 │ 38.7 0.7 1.6
4 │ 35.7 0.3 7.38025
5 │ 21.9 0.2 3.0012DAMM(x::VecOrMat{<: Real}, p::NTuple{7, Float64})
Calculate respiration as a function of soil temperature (Tₛ) and moisture (θ).
julia> df = DAMMfdata(100) # generates a fake dataset
100×3 DataFrame
Row │ Tₛ θ Rₛ
│ Float64 Float64 Float64
─────┼─────────────────────────────
1 │ 15.5 0.3 1.72216
2 │ 22.3 0.6 1.8213
⋮ │ ⋮ ⋮ ⋮
99 │ 9.5 0.2 0.223677
100 │ 6.6 0.6 0.730627
julia> fp # parameters: αₛₓ, Eaₛₓ, kMₛₓ, kMₒ₂, Sxₜₒₜ, Q10kM
(1.0e9, 64.0, 3.46e-8, 0.002, 0.7, 0.02, 1.0)
julia> DAMM(hcat(df.Tₛ, df.θ), fp) # μmolCO₂ m⁻² s⁻¹
100-element Vector{Float64}:
6.023429035220588
0.9298933641647085
⋮
0.8444248717855868
3.805243237387702DAMMfit(x::VecOrMat{<: Real}, Rₛ::Vector{Float64}, poro_val::Float64)
fit the DAMM model parameters to data.
julia> df = DAMMfdata(100) # generates a fake dataset
100×3 DataFrame
Row │ Tₛ θ Rₛ
│ Float64 Float64 Float64
─────┼─────────────────────────────
1 │ 27.1 0.3 4.345
2 │ 38.7 0.6 12.0106
⋮ │ ⋮ ⋮ ⋮
99 │ 18.6 0.5 0.894257
100 │ 19.4 0.4 3.79532
julia> p = DAMMfit(hcat(df.Tₛ, df.θ), df.Rₛ, 0.7)
(2.034002955272664e10, 71.65411256289629, 9.903541279858033e-8, 0.003688664956456453, 0.7, 0.02, 1.0)
julia> DAMM(hcat(df.Tₛ, df.θ), p)
100-element Vector{Float64}:
4.233540174412755
10.41149919818871
⋮
1.746141124513421
1.9599317903590014DAMMmat(Tₛ::Array{Float64, 1}, θ::Array{Float64, 1}, R::Array{Float64, 1}, r::Int64)
Generates a matrix of DAMM output for gridded inputs x and y Inputs: soil temperature (Tₛ), soil moisture (θ), respiration (R), resolution (r)
julia> Tₛ = collect(15.0:2.5:40.0)
julia> θ = collect(0.2:0.05:0.7)
julia> R = [1.0, 1.2, 1.5, 2.0, 2.7, 3.8, 4.9, 6.7, 4.1, 2.0, 0.4]
julia> r = 10
julia> poro_val, params, x, y, DAMM_Matrix = DAMMmat(Tₛ, θ, R, r)DAMMmat(Tₛ::Array{Float64, 1}, θ::Array{Float64, 1}, R::Array{Float64, 1}, r::Int64, n::Int64)
Bin data by n quantiles
julia> n = 4
julia> poro_val, Tmed, θmed, Rmed, params, x, y, DAMM_Matrix = DAMMmat(Tₛ, θ, R, r, n)DAMMplot(Tₛ::Array{Float64, 1}, θ::Array{Float64, 1}, R::Array{Float64, 1}, r::Int64)
Plot scatter of data and fitted DAMM surface
julia> df = DAMMfdata(100)
100×3 DataFrame
Row │ Tₛ θ Rₛ
│ Float64 Float64 Float64
─────┼────────────────────────────
1 │ 31.8 0.3 6.54735
2 │ 14.6 0.3 3.49235
⋮ │ ⋮ ⋮ ⋮
99 │ 17.2 0.4 0.880441
100 │ 5.2 0.2 0.0
julia> r = 50
julia> fig = DAMMplot(df.Tₛ, df.θ, df.Rₛ, r)
qbins(x, y, z, n)
Bins x into n quantiles, each xbin into n quantiles of y, return z quantile
julia> df = DataFrame(x=1:20, y=6:25, z=11:30)
julia> xmed, ymed, zmed = qbins(df.T, df.M, df.R, 3)
xmed = [9, 9, 9, 15, 15, 15, 21, 21, 21]
ymed = [12, 14, 16, 19, 20.5, 22, 25, 27, 29]
zmed = [2, 4, 6, 8.5, 10.5, 15, 17, 19]Issues and pull requests are welcome!