dvdkamp / fsmm

Fuel Stick Moisture Model: Simulates moisture levels and temperature for standardized fuel sticks

fsmm: Fuel Stick Moisture Model

Simulates moisture levels and temperature for standardized fuel sticks

This repository provides code and example forcing data for the model described by:

van der Kamp, D.W., Moore, R.D., and McKendry, I.G. (2017) A model for simulating the moisture content of standardized fuel sticks of various sizes. Agriculture and Forest Meteorology. 236: 123- 134.

For those without access you can download the final manuscript.

It also provides example fuel moisture observations that were used to optimize and evaluate the model

Dependencies

The model requires the package deSolve to be installed:

install.packages("deSolve")
require(deSolve)

Fortran is required

This core of the model is written in Fortran and called by R using the deSolve package. The Shared Object/DLL is generated with R CMD SHLIB and is loaded with dyn.load

As such you will be required to have Fortran installed.

Shortwave partitioning, sun position, and downwelling longwave radiation not yet implemented.

I have yet to implement these three features for this version of the model. As such, you will be required to provide Downwelling direct shortwave radiation (k.down.dir), Downwelling diffuse shortwave radiation (k.down.diff), Sun Altitude (Alt_s), Sun Azimuth (Azi_s), and Downwelling longwave radiation (L_dn) to the model yourself (see details on the fsmm.model.input argument in next section).

The model will be updated soon with these additional features, at which point you will only need to provide total shortwave radiation.

See section A.2 of van der kamp et al (2017) for details on these features.

Running the model

Running run.fsmm() will run the model with default arguments and produce modelled fuel moisture and temperature.

Required arguments for run.fsmm

• fsmm.model.input: a data.frame containing the required forcing variables. The default is to load the example forcing data.frame: data/fsmm.example.forcing.vars.csv. This dataframe has to have the following column names:
  • Alt_s: Sun Altitude (rad)
  • Azi_s: Sun Azimuth. (0 rad at noon, Westerly directions: 0-pi rad Easterly directions: pi to 2pi rad)
    • note: For now, the user will need to calculate these sun position variables themselves. the packages oce and suncalc can help here.
  • k.down.dir: Downwelling direct shortwave radiation (Wm^-2)
  • k.down.diff: Downwelling diffuse shortwave radiation (Wm^-2)
  • temp: Air temperature (^oC)
  • L_dn: Downwelling longwave radiation (Wm^-2)
  • rh: Relative humidity (%)
  • wind.speed: Wind speed (m/s)
  • precip: Precipitation (mm)
• fsmm.model.pars: a vector storing the model parameter values. The default is the optimal parameter values for site "BC1" listed in Table 1 of van der Kamp et al. (2017). The parameters need to be provided in the following order:
  • A : constant for EMC equation
  • B : constant for EMC equation
  • K_s: Diffusivity (m^2 s^-1)
  • M_MAX: Maximum allowable moisture content (fraction)
  • DR_O: volume fraction of the stick taken up by the outer layer
• RHO_S: stick density (kg m-3). default value is 400 kg m-3.
• r: radius of stick (m). Default value is 0.0065 m.
• L: length of stick (m). Deafult value is 0.41 m.
• m_i: initial moisture content of stick (weight fraction). Default value is 0.1.

Output

A matrix with seven columns:

• time Model time (sec)
• fuel.temp.mod.outer Temperature of the outer layer (^oC)
• fuel.mois.mod.outer: Moisture of the outer layer (weight fraction)
• fuel.mois.mod.core: Moisture of the stick core (weight fraction)
• fuel.temp.mod.core: Temperature of the stick core (^oC)
• fuel.mois.mod: Average moisture of the entire stick (weight fraction)
• fuel.temp.mod: Average temperature of the entire stick (^oC)

Example fuel moisture observations

The csv file: data/fsmm.example.obs.fuel.moisture.csv contains hourly observations of fuel moisture for two sites: sites "BC1" and "BC2" mentioned in van der Kamp et al (2017)