Jojo Ono Jojo Ono - 25 days ago 14
R Question

Assigning location values based on date and time of a second data set

I have 2 data frames: 1 is a sequence of GPS locations with associated date-times (POSIXct)

GPS <- data.frame(Lat=c(58.65209, 58.65183, 58.65142, 58.65129, 58.65126, 58.65124, 58.65122, 58.65119, 58.65117, 58.65115),
Lon=c(-3.178559, -3.177934, -3.177277, -3.177536, -3.177494, -3.177713, -3.177806, -3.177899, -3.177991, -3.178084),
datetime=c("2016-10-01 16:23:59 GMT", "2016-10-01 16:31:59 GMT", "2016-10-01 16:39:59 GMT", "2016-10-01 16:47:59 GMT", "2016-10-01 16:55:59 GMT", "2016-10-01 17:03:59 GMT", "2016-10-01 17:11:59 GMT", "2016-10-01 17:19:59 GMT", "2016-10-01 17:27:59 GMT", "2016-10-01 17:35:59 GMT"))

GPS$datetime <- as.POSIXct(as.character(GPS$datetime))


and another is a sequence of depths with associated date-times (POSIXct).

DEPTH <- data.frame(Depth=c(0.0, 0.1, 0.0, 0.0, 0.1, 1.5, 1.7, 1.7, 1.4, 1.5, 1.8, 2.1, 2.3, 1.7, 2.0, 2.6, 2.2, 2.1, 3.4, 3.3),
datetime=c("2016-10-01 16:22:56 GMT", "2016-10-01 16:23:06 GMT", "2016-10-01 16:23:16 GMT", "2016-10-01 16:23:59 GMT", "2016-10-01 16:24:52 GMT", "2016-10-01 16:25:24 GMT", "2016-10-01 16:32:40 GMT", "2016-10-01 16:32:51 GMT", "2016-10-01 18:45:30 GMT", "2016-10-01 18:45:40 GMT", "2016-10-01 18:46:54 GMT", "2016-10-01 18:47:04 GMT", "2016-10-01 18:47:14 GMT", "2016-10-01 18:47:25 GMT", "2016-10-01 18:51:03 GMT", "2016-10-01 18:51:14 GMT", "2016-10-01 18:51:24 GMT", "2016-10-01 18:54:11 GMT", "2016-10-01 18:54:21 GMT", "2016-10-01 18:54:32 GMT"))

DEPTH$datetime <- as.POSIXct(as.character(DEPTH$datetime))


For each depth location I want to assign a location (latitude and longitude) based on when the interpolated track from the location data frame suggests it should be i.e. if the locations go from point A to point B at which point along that line does the depth data lie, assuming a uniform speed between points, given it's date-time.

the final product would be 2 vectors in the data-frame which assign each depth value with a latitude and a longitude.

Thank you.

Answer

Assuming constant velocity (uniform speed and constant direction) between successive way-points in the GPS data, we can perform linear interpolation separately for Lat and Lon using approx from stats:

DEPTH$Lat <- approx(x=GPS$datetime, y=GPS$Lat, xout=DEPTH$datetime, method="linear")$y
DEPTH$Lon <- approx(x=GPS$datetime, y=GPS$Lon, xout=DEPTH$datetime, method="linear")$y

In this usage, interpolated values for datetime in DEPTH outside of the range of datetime in GPS are assigned NA. See ?approx for other ways to dealing with interpolating outside the range of the input.

Using your data, the result is:

print(DEPTH)
##   Depth            datetime      Lat       Lon
##1    0.0 2016-10-01 16:22:56       NA        NA
##2    0.1 2016-10-01 16:23:06       NA        NA
##3    0.0 2016-10-01 16:23:16       NA        NA
##4    0.0 2016-10-01 16:23:59 58.65209 -3.178559
##5    0.1 2016-10-01 16:24:52 58.65206 -3.178490
##6    1.5 2016-10-01 16:25:24 58.65204 -3.178448
##7    1.7 2016-10-01 16:32:40 58.65179 -3.177878
##8    1.7 2016-10-01 16:32:51 58.65179 -3.177863
##9    1.4 2016-10-01 18:45:30       NA        NA
##10   1.5 2016-10-01 18:45:40       NA        NA
##11   1.8 2016-10-01 18:46:54       NA        NA
##12   2.1 2016-10-01 18:47:04       NA        NA
##13   2.3 2016-10-01 18:47:14       NA        NA
##14   1.7 2016-10-01 18:47:25       NA        NA
##15   2.0 2016-10-01 18:51:03       NA        NA
##16   2.6 2016-10-01 18:51:14       NA        NA
##17   2.2 2016-10-01 18:51:24       NA        NA
##18   2.1 2016-10-01 18:54:11       NA        NA
##19   3.4 2016-10-01 18:54:21       NA        NA
##20   3.3 2016-10-01 18:54:32       NA        NA
Comments