modelsnowdepth: Runs spatial snow model

In ilyamaclean/microsnow: Below or above canopy microclimate modelling with R in snowy environments

Runs spatial snow model

Description

modelsnowdepth runs the spatial snow model to derive an array of snow depths and temperatures for each time increment in weather and over each grid cell of dtm

Usage

modelsnowdepth(
  weather,
  precd,
  snowdepth,
  dtm,
  pai,
  hgt,
  STparams,
  meltfact = NA,
  slr = NA,
  apr = NA,
  plai = 0.3,
  x = 1,
  lat = NA,
  long = NA,
  snowenv = "Taiga",
  tpi_radius = 200,
  tfact = 10,
  snowem = 0.99,
  zmin = 0.002,
  umin = 0.5,
  astc = 1.5,
  spatialmelt = FALSE,
  Sh = 6.3,
  zu = 2,
  xyf = NA,
  initdepth = 0,
  out = "hourly",
  clump = 0.2
)

Arguments

weather	a data.frame of weather variables (see details).
precd	a vector of daily precipitation (mm).
snowdepth	a numeric vector of length equal to the number of timesteps (e.g. nrow(weather)) representing average snowdepth, in cm, across the scene as e.g. returned by runNMRSnow() or spline interpolated from measurments.
dtm	a SpatRast of elevations (m). the x and y dimensions of the raster must also be in metres
pai	a single numeric value, SpatRast or array of plant area index values
hgt	a SpatRast of vegetation heights
STparams	snow temperature model coefficients as derived by fitsnowtemp()
meltfact	snow melt coefficient as returned by fitsnowtemp or getmeltf. Derived using getmeltf is not supplied.
slr	an optional SpatRast of slope values (Radians). Calculated from dtm if not supplied, but outer cells will be NA.
apr	an optional SpatRast of aspect values (Radians). Calculated from dtm if not supplied, but outer cells will be NA.
plai	a single numeric value, SpatRast or array of the proprotion of plant area index values that are leaves
x	optional single numeric value, SpatRast, matrix or array of leaf distribution coefficients
lat	latitude of location (decimal degrees). Derived from dtm is not supplied, so coordinate reference of system of dtm must be defined.
long	longitude of location (decimal degrees). Derived from dtm is not supplied, so coordinate reference of system of dtm must be defined.
snowenv	one of Alpine, Maritime, Prairie, Taiga or Tundra (see details)
tpi_radius	radius for applying topographic positioning index when distributing snow as returned by tpiradius()
tfact	single numeric value specifying sensitivity of snow distribution to topographic positioning index
snowem	optionally, numeric value of snow emissivity
zmin	optionally, numeric value of roughness length for momentum transfer of snow surface without vegetation protruding (m)
umin	optionally, numeric value indicating minimum wind speed used in conductivity calculations (m/s)
astc	optionally, numeric value indicating the temperature at which precipitation falls as snow (deg C)
spatialmelt	optional logical indicating whether or not snow melt varies spatially (e.g. greater on sun-facing slopes). Model takes longer to run if TRUE
Sh	optionally, branch snow load coefficient (kg/m^2). 6.6 for pine and 5.9 kg for spruce.
zu	height above ground of wind speed measurement in weather (m)
xyf	number of grid cells over which to smooth vertical wind height profile
initdepth	single numeric value or matrix of initial snow depths at start of model run
out	optional variable indicating whether to return hourly or daily snow depths (default is hourly)
clump	a single numeric value or array of values between 0 and 1 indicating the fraction of radiation passing through larger gaps in the canopy, ( see microclimf::clumpestimate())

Details

#' @details The format and and units of weather must follow that in the example dataset climdata. The paramater snowenv is used to compute snow density following Sturm et al (2010) J Hydrometeorology 11: 1380-1393. The leaf distribution angle coefficient is the ratio of vertical to horizontal projections of leaf foliage (~1 for decidious woodland).

Value

a list of the following:

gsnowdepth array of predicted ground snow depth (cm) across the scene

canswe array of predicted canopy snow water equivalent (mm / m^2) across the scene

snowtempG array of predicted ground snow surface temperature (deg C) across the scene

Examples

# Run snow model with defaults and inbuilt datasets and model coefficients (takes 20 seconds)
snd<-nowdepth$snowdepth # snow depth vector
mout1<-modelsnowdepth(climdata,precd,snd,dtm,pai,hgt,STparams,SDparams$meltfact)
# Run snow model with defaults, but allowing spatially variable snow melt (takes ~100 seconds)
mout2<-modelsnowdepth(climdata,precd,snd,dtm,pai,hgt,STparams,SDparams$meltfact,spatialmelt = TRUE)
# Compare results
msnowdepth1<-apply(mout1$gsnowdepth,c(1,2),mean)
msnowdepth2<-apply(mout2$gsnowdepth,c(1,2),mean)
snowdif<-msnowdepth2-msnowdepth1
par(mfrow=c(2,2))
plot(rast(msnowdepth1))
plot(rast(msnowdepth2))
plot(rast(snowdif))

ilyamaclean/microsnow documentation built on April 7, 2023, 8:55 a.m.