0

I am working to test whether points are clustering as a result of density of those points or irregardless of density, in this case clustering (broods of juvenile shorebirds).

I have location data in discrete time steps for multiple individuals. I am hoping to calculate the average distance between all combinations of two points at each time step. What tools are available to do this?

I have tried to accomplish this within the dplyr package, but have not had much luck. I know this will likely create a large matrix of these points, hence why I would want to calculate a statistic such as average from it.

my data look like this: (with Lat and Long in UTM's, one location per brood, per date)

> dput(dista)
structure(list(Plot = c("S", "S", "S", "S", "S", "S", "S", "S", 
"S", "S", "S", "S", "S", "S", "S", "S", "S", "S", "S", "S", "S", 
"S", "S", "S", "S", "S", "S"), Brood = c("GB03", "GB03", "GB03", 
"GN01", "GN01", "GN01", "GN01", "GN01", "GN01", "GN01", "GN02", 
"GN02", "GN02", "GN02", "GN02", "GN02", "GN02", "GN02", "GN02", 
"GN02", "GN02", "GN02", "GN02", "GN02", "GN02", "GN03", "GN03"
), Day = c("6/17/2019", "6/19/2019", "6/22/2019", "6/3/2019", 
"6/5/2019", "6/7/2019", "6/8/2019", "6/10/2019", "6/12/2019", 
"6/14/2019", "6/3/2019", "6/7/2019", "6/8/2019", "6/12/2019", 
"6/14/2019", "6/17/2019", "6/19/2019", "6/22/2019", "6/24/2019", 
"6/26/2019", "6/28/2019", "7/1/2019", "7/3/2019", "7/8/2019", 
"7/10/2019", "6/18/2019", "6/20/2019"), Lat = c(601701.2, 601640.2, 
601806.2, 602398.8, 602457.3, 602447.8, 602529.2, 602467.4, 602420.4, 
602422.3, 602124.9, 601706.6, 601791.2, 601570.3, 602221.4, 602867.3, 
602069.7, 602065.6, 601995.5, 602603.3, 602034.2, 601878.1, 602246.5, 
602452.9, 602203.3, 601476.8, 601719.9), Long = c(6778175L, 6778016L, 
6778143L, 6778541L, 6778558L, 6778629L, 6778700L, 6778526L, 6778538L, 
6778440L, 6778510L, 6778172L, 6778296L, 6777853L, 6778437L, 6778700L, 
6778513L, 6778465L, 6778471L, 6780454L, 6777941L, 6778173L, 6777624L, 
6778407L, 6778204L, 6777993L, 6778037L)), class = "data.frame", row.names = c(NA, 
-27L))

I want to: 1) produce a measure of distance between all individuals at each time step 2) be able to eventually plot these values to test for effects of density on these distances

Luke Wilde
  • 1
  • 1
  • 1
    Please share your data and the code you've tried. You can get some hints on how to do that [here](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example). – Roman Luštrik Sep 02 '19 at 17:04
  • You can very likely use [`proxy`](http://cran.r-project.org/package=proxy). – Alexis Sep 02 '19 at 17:08
  • The `dput` function is a great way to share data - it makes it easy for anyone to paste your data into their R session. In this case, you could use `dput(dista)`. – shwan Sep 02 '19 at 19:18
  • Good thinking, Roman Luštrik and shawn! I have attached my data. As far as the code goes, I have mainly worked to create Kerned UDs and estimate area from that. This is not where I want to keep the computation as it drastically oversimplifies the query. – Luke Wilde Sep 03 '19 at 20:53

0 Answers0