Getting distance between vectors of longitude and latitude points

Question

I have a data set that looks like this:

structure(list(Date2 = structure(c(18428, 18438, 18428, 18438, 
18428, 18438, 18428, 18438, 18428, 18438, 18428, 18438, 18428, 
18438, 18428, 18438, 18428, 18438, 18428, 18438, 18428, 18438, 
18428, 18438, 18428, 18438, 18428, 18438, 18428, 18438, 18428, 
18438, 18428, 18438, 18428, 18438, 18428, 18438, 18428, 18438, 
18428, 18438, 18428, 18438), class = "Date"), Fish_ID = c("Fork1", 
"Fork1", "Fork10", "Fork10", "Fork12", "Fork12", "Fork13", "Fork13", 
"Fork14", "Fork14", "Fork15", "Fork15", "Fork16", "Fork16", "Fork17", 
"Fork17", "Fork18", "Fork18", "Fork19", "Fork19", "Fork2", "Fork2", 
"Fork20", "Fork20", "Fork21", "Fork21", "Fork22", "Fork22", "Fork23", 
"Fork23", "Fork3", "Fork3", "Fork4", "Fork4", "Fork5", "Fork5", 
"Fork6", "Fork6", "Fork7", "Fork7", "Fork8", "Fork8", "Fork9", 
"Fork9"), Lat2 = c(32.9394, NA, 32.92935, NA, NA, 32.9047333333333, 
NA, 32.9093833333333, NA, 32.9509833333333, 32.9160666666667, 
NA, NA, 32.9074333333333, NA, 32.9029, NA, 32.90775, NA, 32.9094, 
NA, NA, 32.9455166666667, 32.9459166666667, 32.9431, 32.9437666666667, 
32.90365, 32.9044333333333, 32.9056166666667, 32.90585, NA, 32.9475333333333, 
32.94325, NA, 32.9288833333333, NA, NA, NA, 32.9297, NA, NA, 
NA, 32.9303, NA), Long2 = c(-95.6334, NA, -95.6406, NA, NA, -95.6531666666667, 
NA, -95.6486, NA, -95.6252333333333, -95.648, NA, NA, -95.6391166666667, 
NA, -95.64155, NA, -95.6393666666667, NA, -95.63895, NA, NA, 
-95.6391166666667, -95.6389333333333, -95.6365, -95.6401333333333, 
-95.6535666666667, -95.6532833333333, -95.6560333333333, -95.6575166666667, 
NA, -95.63015, -95.6334333333333, NA, -95.6395, NA, NA, NA, -95.6398833333333, 
NA, NA, NA, -95.6425166666667, NA), lag.Lat2 = c(NA, 32.9394, 
NA, 32.92935, NA, NA, NA, NA, NA, NA, NA, 32.9160666666667, NA, 
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 32.9455166666667, NA, 
32.9431, NA, 32.90365, NA, 32.9056166666667, NA, NA, NA, 32.94325, 
NA, 32.9288833333333, NA, NA, NA, 32.9297, NA, NA, NA, 32.9303
), lag.Long2 = c(NA, -95.6334, NA, -95.6406, NA, NA, NA, NA, 
NA, NA, NA, -95.648, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 
NA, -95.6391166666667, NA, -95.6365, NA, -95.6535666666667, NA, 
-95.6560333333333, NA, NA, NA, -95.6334333333333, NA, -95.6395, 
NA, NA, NA, -95.6398833333333, NA, NA, NA, -95.6425166666667)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -44L), groups = structure(list(
    Fish_ID = c("Fork1", "Fork10", "Fork12", "Fork13", "Fork14", 
    "Fork15", "Fork16", "Fork17", "Fork18", "Fork19", "Fork2", 
    "Fork20", "Fork21", "Fork22", "Fork23", "Fork3", "Fork4", 
    "Fork5", "Fork6", "Fork7", "Fork8", "Fork9"), .rows = structure(list(
        1:2, 3:4, 5:6, 7:8, 9:10, 11:12, 13:14, 15:16, 17:18, 
        19:20, 21:22, 23:24, 25:26, 27:28, 29:30, 31:32, 33:34, 
        35:36, 37:38, 39:40, 41:42, 43:44), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -22L), .drop = TRUE))

Lat2 and Long2 are the locations of each fish and lag values are for the location of each fish the prior time they were located.

I am trying to calculate the distance between each Long2 Lat2 value and each lag.Long2 lag.Long2 value so that I can calculate the distance traveled from the last time each fish was located. I know how to do this by hand for each one using the geosphere package, but I'm wondering if there is a for loop I could write to do this calculation for each individual fish so I could automate the process?

Thanks!

Are you looking for `rdist()` in the `fields` package? – John Polo Aug 08 '22 at 17:21 — John Polo, Aug 08 '22 at 17:21
can you post the code that does this for one row? – sjp Aug 08 '22 at 17:27 — sjp, Aug 08 '22 at 17:27

Chris · Accepted Answer · 2022-08-08T21:05:05.107

0

using your data, we're looking for fish that moved, fishx and fishy, as matrix, with long before lat, so perhaps:

complete <- which(complete.cases(fish_df)==TRUE)
fishx <- as.matrix(fish_df[complete, 4:3])
fishx
       Long2     Lat2
24 -95.63893 32.94592
26 -95.64013 32.94377
28 -95.65328 32.90443
30 -95.65752 32.90585
fishy <- as.matrix(fish_df[complete, 6:5])
geosphere::distm(fishx, fishy, fun = geosphere::distGeo)
           [,1]      [,2]       [,3]      [,4]
[1,]   47.55876  386.4671 4883.23872 4746.9509
[2,]  216.11520  347.7147 4623.08009 4484.7115
[3,] 4745.03178 4566.5523   90.82777  288.8099
[4,] 4723.80758 4574.9848  442.81633  141.1616
diag(geosphere::distm(fishx, fishy, fun = geosphere::distGeo))
[1]  47.55876 347.71468  90.82777 141.16164
# presumably in meters

You know your study area so this may or may not be correct... Which is to say, no need for loop, just get your matrices.

edited Aug 08 '22 at 21:05

answered Aug 08 '22 at 20:59

Chris

1,647
1
18
25

thanks for showing me that with the diag. This is correct to get what I want, however I want to keep the NAs in there. With this code: `res <- with(df, distm(cbind(Long2, Lat2), cbind(lag.Long2, lag.Lat2), distHaversine)) diag(res)` I'll get what I'm after. Thanks! – David Smith Aug 09 '22 at 13:25
so `diag(distm(as.matrix(fish_df[, 4:3]), as.matrix(fish_df[, 6:5]), distHaversine))`, though I suppose one should check effect of NA on distHaversine...I went with complete.cases so I could see I was getting somewhere and was reasonably mapping to inputs, producing reasonable outputs, which of course it was. Attempting mapping of ocean fishing grounds environmentals from remote sensing myself. – Chris Aug 09 '22 at 14:39

Getting distance between vectors of longitude and latitude points

1 Answers1