Skip to contents

Simulate interpolated flow fields from tiltmeter velocity data

Source: R/simulate_flowfields.R
simulate_flowfields.Rd

Interpolates eastward and northward velocity components from tiltmeter observations onto a regular spatial grid for each time step. The function aggregates tiltmeter observations to a user-defined temporal resolution, converts velocities from cm/s to m/s, creates a buffered spatial domain around the instruments, and uses inverse distance weighting to generate gridded flow fields.

Usage

simulate_flowfields(
  tiltmeters,
  gbr_buffer,
  crs_m = 32755,
  grid_res_m = 10,
  buffer_m = 250,
  time_unit = "5 minutes",
  idp = 2
)

Arguments

tiltmeters

An sf object or data frame containing tiltmeter observations. Must contain columns time, inst, lat, lon, velocity_e_cm_s, velocity_n_cm_s, and speed_cm_s.

gbr_buffer

An sf polygon object defining the spatial mask or reef buffer used to constrain the interpolation domain.

crs_m

Numeric or character. Projected coordinate reference system used for interpolation in metres. Default is 32755.

grid_res_m

Numeric. Grid cell spacing in metres. Default is 10.

buffer_m

Numeric. Buffer distance in metres around the convex hull of tiltmeter locations. Default is 250.

time_unit

Character. Temporal aggregation unit passed to lubridate::floor_date(). Default is "5 minutes".

idp

Numeric. Inverse distance weighting power passed to gstat::idw(). Higher values give more weight to nearby observations. Default is 2.

Value

A named list with:

flow_field

An sf point object in crs_m containing interpolated u_ms, v_ms, speed_ms, and heading_degrees for each grid cell and time step.

flow_field_4326

The same interpolated flow field transformed to EPSG:4326.

tiltmeters_m

Aggregated tiltmeter observations transformed to crs_m.

grid_sf

The interpolation grid as an sf point object in crs_m.

domain_m

The buffered interpolation domain clipped to gbr_buffer.

Details

The function first aggregates tiltmeter measurements by instrument and floored time interval. Eastward and northward velocities are converted from cm/s to m/s. A convex hull is constructed around the tiltmeter locations, buffered by buffer_m, and clipped to gbr_buffer. A regular point grid is then generated over this domain.

For each time step, inverse distance weighting is applied separately to the eastward (u_ms) and northward (v_ms) velocity components. Wind/current speed is then calculated as:

$$speed = \sqrt{u^2 + v^2}$$

Heading is calculated in degrees using:

$$heading = atan2(u, v)$$

and converted to compass-style degrees from 0 to 360.

Time steps with fewer than three tiltmeter observations are skipped.

Examples

if (FALSE) { # \dontrun{
flow_sim <- simulate_flowfields(
  tiltmeters = tiltmeters,
  gbr_buffer = gbr_buffer,
  crs_m = 32755,
  grid_res_m = 10,
  buffer_m = 250,
  time_unit = "5 minutes",
  idp = 2
)

flow_sim$flow_field
flow_sim$flow_field_4326
} # }