In eldarrak/FLightR: Reconstruct Animal Paths from Solar Geolocation Loggers Data

Intigeo tag on a Bar tailed godwit analysis example

GeoLight analysis

Appendix A5 to Rakhimberdiev, E., Senner, N. R., Verhoeven, M. A., Winkler, D. W., Bouten, W. and Piersma T. 2016 Comparing inferences of solar geolocation data against high-precision GPS data: annual movements of a double-tagged Black-Tailed Godwit. - Journal of Avian Biology 000: 000-000.

We used GeoLight 2.01, so you might want to install it if you want to get exactly the same results.

library(devtools)
install_github("SLisovski/GeoLight@Update_2.01")
library(GeoLight)

Download data from GitHub

There are two files available in the directory: .lux and .csv. *.lux is the original file you get from Migrate Technology Ltd. This file does not have defined twilights. In the appendix A4 we already defined twilights with [BAStag package] (https://github.com/SWotherspoon/BAStag) and saved them. You can define twilights by yourself or just download file with predefined format. Let's now assume that you have done the previous step and have a .csv file. Now you can process these data:

download.file(
 "https://raw.githubusercontent.com/eldarrak/FLightR/master/examples/Black-Tailed_Godwit_JAB_example/A3_TAGS_format.csv",
 "A3_TAGS_format.csv")
TAGS.twilights<-read.csv("A3_TAGS_format.csv", stringsAsFactors =F)

Prepare data

TAGS.twilights$light<-exp(TAGS.twilights$light) # this is needed because
                      # we log transformed data in convert.lux.to.tags()

TAGS.twilights$datetime<-as.POSIXct(TAGS.twilights$datetime,
                                    tz="UTC", format="%Y-%m-%dT%T")

# so we have to extract twilights
tw<-TAGS.twilights[TAGS.twilights$twilight>0 &  TAGS.twilights$excluded==0
 & TAGS.twilights$datetime<as.POSIXct("2014-05-18", tz="UTC"),]

# make pairs
tw$datetime[tw$twilight==2]<-tw$datetime[tw$twilight==2]-(300-60)

gl_twl<-data.frame(tFirst=tw$datetime[which(diff(tw$datetime)<20)],
                   tSecond=tw$datetime[which(diff(tw$datetime)<20)+1],
                   type=tw$twilight[which(diff(tw$datetime)<20)])

Calibration

#Calibrate the sun elevation angle with a known location
#designate first twilights for calibration
calib1 <- gl_twl[gl_twl$tFirst<=as.POSIXct("2013-08-05", tz="UTC"),] 
known.coord <- c(5.47, 53.04)

elev <- getElevation(calib1$tFirst,calib1$tSecond,
                     calib1$type,known.coord=known.coord,
                     plot=T) #get sun elevation angle 

Generate the coordinates

crds <- coord(twl=gl_twl, degElevation = elev)

#Add times to GeoLight crds
crds <- data.frame(Time = as.POSIXct(apply(cbind(gl_twl$tFirst,gl_twl$tSecond)
                                           , 1 , mean), tz="UTC", origin="1970-01-01"),
                   Lon  = crds[,1], Lat = crds[,2])

tripMap(crds[,c("Lon", "Lat")], xlim=c(-10, 10), ylim=c(0, 60))

Estimate sites and migration schedules

cL <- changeLight(gl_twl,  quantile = 0.95, days = 3) 

mS <- mergeSites(gl_twl, site = cL$site, degElevation = elev, distThreshold = 400)

siteMap(crds[,c("Lon", "Lat")], mS$site, type = "cross", hull = F, xlim = c(-12, 12), ylim = c(30, 58))

(schedule(tFirst = gl_twl$tFirst, gl_twl$tSecond, site = mS$site))

Result:

  Site             Arrival           Departure
1    a                <NA> 2013-08-20 11:55:25
2    b 2013-08-26 00:10:58 2014-04-20 00:26:08
3    c 2014-04-22 00:17:36 2014-04-28 11:59:32
4    d 2014-05-02 23:45:04                <NA>

crds$Site <- mS$site

save(crds, file="GeoLight_coords.RData")

Now we will proceed to FLightR workflow.

eldarrak/FLightR documentation built on Nov. 10, 2023, 7:53 a.m.