Returns values for the selected quantity from a vertical profile (vp), list, or time series of vertical profiles (vpts). Values are organized per height bin. Values for eta are set to 0, dbz to -Inf and ff, u, v, w, dd to NaN when the sd_vvp for that height bin is below the sd_vvp_threshold().

get_quantity(x, quantity)

# S3 method for vp
get_quantity(x, quantity = "dens")

# S3 method for list
get_quantity(x, quantity = "dens")

# S3 method for vpts
get_quantity(x, quantity = "dens")

Arguments

x

A vp, list of vp or vpts object.

quantity

Character. A (case sensitive) profile quantity, one of:

  • height: Height bin (lower bound) in m above sea level.

  • u: Speed component west to east in m/s.

  • v: Speed component north to south in m/s.

  • w: Vertical speed (unreliable!) in m/s.

  • ff: Horizontal speed in m/s.

  • dd: Direction in degrees clockwise from north.

  • sd_vvp: VVP radial velocity standard deviation in m/s.

  • gap: Angular data gap detected in T/F.

  • dbz: Animal reflectivity factor in dBZ.

  • eta: Animal reflectivity in cm^2/km^3.

  • dens: Animal density in animals/km^3.

  • DBZH: Total reflectivity factor (bio + meteo scattering) in dBZ.

  • n: Number of data points used for the ground speed estimates (quantities u, v, w, ff, dd).

  • n_all: Number of data points used for the radial velocity standard deviation estimate (quantity sd_vvp).

  • n_dbz: Number of data points used for reflectivity-based estimates (quantities dbz, eta, dens).

  • n_dbz_all: Number of data points used for the total reflectivity estimate (quantity DBZH).

  • attributes: List of the vertical profile's what, where and how attributes.

Value

For a vp object: a named (height bin) vector with values for the selected quantity.

For a list object: a list of named (height bin) vectors with values for the selected quantity.

For a vpts object: a (height bin * datetime) matrix with values for the selected quantity.

See also

Examples

# Load the example vertical profile vp <- example_vp # Extract the quantity animal density (dens) get_quantity(vp, "dens")
#> 0 200 400 600 800 1000 #> NA 85.0995178 110.4298782 91.7822418 34.5677528 20.5497875 #> 1200 1400 1600 1800 2000 2200 #> 22.9609985 19.7929668 19.7772617 17.3985100 11.9922190 7.2263165 #> 2400 2600 2800 3000 3200 3400 #> 3.0735207 0.0000000 0.7724188 0.0000000 0.0000000 0.0000000 #> 3600 3800 4000 4200 4400 4600 #> 0.0000000 0.0000000 0.0000000 0.0000000 0.0000000 0.0000000 #> 4800 #> 0.0000000
# Load the example time series of vertical profiles vpts <- example_vpts # Extract the quantity horizontal speed (ff) and show the first two datetimes get_quantity(vpts, "ff")[,1:2]
#> 2016-09-01 00:02:00 2016-09-01 00:06:00 #> 0 NaN NaN #> 200 NaN NaN #> 400 NaN NaN #> 600 5.65 5.23 #> 800 5.07 5.64 #> 1000 5.45 5.39 #> 1200 6.02 6.20 #> 1400 5.96 5.81 #> 1600 4.82 4.63 #> 1800 4.61 4.07 #> 2000 NaN 2.74 #> 2200 NaN 2.28 #> 2400 NaN 0.73 #> 2600 NaN 0.74 #> 2800 NaN NaN #> 3000 NaN NaN #> 3200 NaN NaN #> 3400 NaN 1.11 #> 3600 NaN 1.17 #> 3800 NaN NaN #> 4000 NaN NaN #> 4200 NaN NaN #> 4400 NaN NaN #> 4600 NaN NaN #> 4800 NaN NaN