Jump to content
Ornithology Exchange
Sign in to follow this  

About This Group

This group provides support for those using geolocator technology to study animal movement and behavior.

  1. What's new in this group
  2. Bonjour, I am working with geolocators data collected with Intigeo C65 devices on seabirds breeding in the Arctic (higher red dot on map1) and migrating in southern Atlantic in winter. Parts of the migration occurs during spring and fall equinoxes. I used a rooftop cablibration made in southern Quebec (lower red dot on map1) for about 14 days and I used the findHEZenith function to calculate a new zenith angle on the wintering area. Then I set a list of Zenith changing depending on the period of the year. All my script is running, but for some birds, I have a strange movement up-north after a first staging period. When I create the initial path (thresholdPath function), everything seems to be okay, but when I run the estelleMetropolis function in 3 steps (burn-in, tuning, final runs), no matter how many iterations I do and no matter how many chains I set, the up-north movement appears. On the path on map2, the blue part is ± 10 days around equinoxes, gray is ± 15 days and black ± 20 days. I tried changing all the parameters one by one, but the problem persists. Any idea what could cause this issue? Merci, Yannick
  3. Hi Melina, please make sure that there are no NA´s in the date: sum(is.na(data$Date)) If this is >0 than this is the fault and you need to remove these rows. Cheers, Simeon
  4. Hi everyone! I am running the preprocessLight of my geolocators data and I have the following error in some of my devices: Error in seq.default(as.numeric(tmin) + dt/2, as.numeric(tmax) - dt/2, : 'from' must be of length 1 My data spans not for more than one year but it includes different years (2019 and 2020). I have tried to subset the .lig original data so that it includes only one year, but it doesn't work! Any idea? Someone with the same problem? Thanks !!! MElina
  5. Hi Eli, Everything works for me at the moment, but after updating R and ggmap to the newest versions. Stamenmaps also fail a lot, and thus are hard to use within a function. I will wait for something more reliable before changing a function myself, but everyone is welcome to write a map_FLightR_stamenmap() Cheers, Eldar
  6. I went around the block with google maps for a couple of hours and could not get past this cryptic error message when trying to download a simple example map: ggmap::register_google(key = APIkey) map <- get_map(location = "texas", zoom = 6, source = "stamen") Source : https://maps.googleapis.com/maps/api/staticmap?center=texas&zoom=6&size=640x640&scale=2&maptype=terrain&key=xxx Error in aperm.default(map, c(2, 1, 3)) : invalid first argument, must be an array In addition: Warning message: In get_googlemap(center = location, zoom = zoom, filename = filename) : HTTP 400 Bad Request Others have had this issue and some have solved it--there are some suggestions on stackoverflow--but not me. I tried a bunch of enabling and management options for static maps API, but nothing worked. Maybe something has changed recently with google or ggmap?? Anyway, I'm now building my maps from scratch using get_stamenmap(). Might be a better future option for map.FLightR.ggmap().
  7. Hi, I am trying to use TwGeos::preprocessLight() to identify twilights in light data collected by Lotek LAT280 tags. The below is the log of the "Light Intensity" variable. This approach doesn't seem to give great definition between the dark and light sections and makes picking a threshold challenging. Does anyone have any suggestions for a transformation, a threshold, or an alternative approach with the data from this tag? Thanks!
  8. Hi, I'm analysing some geolocator data from a common sandpiper using the geolight package. When I run my analysis I get a weird migration pattern, the individual appears to stop in two locations (one of which is in the sea) before settling in West Africa. I'm not sure whether these are true movements from west to east or if something else is going on. Does anyone have any suggestions? These movements are not around the equinoxes either. The output of siteMap() and schedule() are attached. Thanks, Thomas ADplot.tiff AD_migration_schedule.csv
  9. Hi Ana, these cut-off probs just measure how many particles movet between time t and t+1. 0.1 will say that there was movement between two periods if at least 10% of particles moved >25 km. 0.5 would mean 50% chance. Generally you better make it higher than lower. There is so far no established approach on how to select a proper value here. Hope this helps, Eldar
  10. Hi all, I estimated stopover locations by using 0.1, 0.2, and 0.3 cutt-off prob. Results are quite consistent specially during the first part of migration. Can the 0.1, 0.2, and 0.3 values be translated into something more "meaningful"? Like km? I'm not quite sure how to explain these three values. Many thanks, Ana
  11. Finally, an almost complete user`s guide for geolocator analyses. Check out the paper: https://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2656.13036 As well as the online manual: https://geolocationmanual.vogelwarte.ch/ Lisovski_et_al-2019-JAnimEcol.pdf
  12. So you can install it directly from there with the install.packages(). 0.4.9 version introduces FLightR2Movebank( ) function that saves result formovebank upload. The latest version is on GitHub, as always.
  13. Sorry - I am not sure what is happening.. are you working on a mac? If not try gr.Device = "default"
  14. Thanks for your help, Eldar! Distance does seem to be key. Increasing max distance in particle.filter has improved the return track. Now working on adjusting mean and sd. Forgive a beginner question, but is there an easy way to estimate appropriate values? I have the result of a particle.filter run that produced an OK-looking track, but not quite sure what I should use. Should I include all movements in my estimates? These are birds that remain in a relatively small area for most of the year, but travel long distances quickly during migration...
  15. Hi Simeon, Thank you so much for responding. I checked my NA's and didn't get any. summary(raw) Date Light Min. :2013-02-22 00:00:02 Min. : 0.0002 1st Qu.:2013-04-28 17:58:47 1st Qu.: 0.0002 Median :2013-07-03 11:57:32 Median : 6.4847 Mean :2013-07-03 11:57:32 Mean : 5.1559 3rd Qu.:2013-09-07 05:56:17 3rd Qu.: 8.8506 Max. :2013-11-11 23:55:02 Max. :11.2176 I set the threshold to -.6, .6, 1, 2. I went through the steps to select the beginning, ending date and the darkest spot in the no light area in order for the package to select the sunrise and sunsets. I accepted all the other screens and then the error keeps happening particularly when I hit (a)accept for stage depicted in the attached image.
  16. Hi Jenny, 1. You could try increasing the distance they are allowed to fly. The maximum distance can be fixed in particle.filter and the mean and sd in make.prerun.object. 2. Try running first without any spatial constrains. It sometimes happen that particle filter just cannot find a proper solution if bir migrated to far every day for a long period. Hope this helps, Eldar
  17. Hi, all I'm using FLightR 0.4.9 to track migration of shorebirds. Tags were deployed in Alaska and recovered at the same site a year later. FLightR gives me nice-looking tracks for the breeding season in Alaska, fall migration to Chile, and the non-breeding season in Chile. But the track for spring migration is consistently problematic -- instead of returning north to Alaska, it heads south toward Antarctica. Has anyone experienced something similar? Any ideas for what I can do? Calibration periods are from the study site (lon = -150.725, lat = 61.272): one from the beginning of the study (shortly after tag deployment) and another from the end (shortly before tag recovery). Both verified by plot_slopes_by_location. > Calibration.periods <- data.frame(calibration.start=as.POSIXct(c("2009-05-17", "2010-05-10")), calibration.stop=as.POSIXct(c("2009-06-15", NA)),lon=-150.725, lat=61.272) > print(Calibration.periods) calibration.start calibration.stop lon lat 1 2009-05-17 2009-06-15 -150.725 61.272 2 2010-05-10 <NA> -150.725 61.272 Calibrating from pre-deployment data instead yields similar results. For the particle filter run I've tried known.last=TRUE and known.last=FALSE. Changing behavioral masks, outlier check, nParticles, etc. also do not seem to help. > all.in <- make.prerun.object(Proc.data, Grid, start=c(-150.725, 61.272), Calibration=Calibration) > nParticles=1e6 > Result <- run.particle.filter(all.in, threads=-1, nParticles=nParticles, known.last=TRUE, check.outliers=TRUE) Since I'm using the correct lon/lat at the end of the track to calibrate, why might the output show my end coordinates so far away from there? Jenny
  18. Hi Diane, have you plotted your light data to make sure that the threshold is within the range of measured light. And have you got any NA's in your time column (check for sum(is.na(raw$Date))). Simeon
  19. Hi I'm working on my thesis playing with the GCSP tag data sets, trying to process them with GeoLocTools and FLightR. I am coming up on an error message when I selecting my twilight threshold. I have tried changing the threshold a variety of different times and still receive the same error message. > threshold <- 0.6 > twl <- preprocessLight(raw, + threshold = threshold, + offset = offset, + lmax = 20, # max. light value (adjust if contrast between night and day is weak) + gr.Device = "x11") Waiting for input Error in class(xx) <- cl : attempt to set an attribute on NULL Any thoughts or suggestion would be greatly appreciated. Thank you so much Diane Nelson
  20. That is correct, median deviation for thresholdPath, since it ignores the error structure and zero devotion for the Estelle model. Running calibration during the equinox ist actually a good thing, if you have enough time around the equinox. Given the structure of errors around the equinox, estimates of latitude are very sensitive to zenith angles (sun elevation angles). Hope that helps, Simeon
  21. Hi! I am running my data with SGAT, but I am not sure which zenith should I use in the thresholdPath and which in the Estelle model. I think I should use the median deviation zenith in the thresholdPath and the zero deviation in the Estelle? Also, which could it be the effect of running my calibration during the spring equinox? Thank you in advance!!! Melina
  22. This is strange. I have just tried installing from GitHub and got the 0.4.9... It is not on CRAN yet, so do not try there. Eldar
  23. Hi Eldar, I just tried updating to FLightR version 0.4.9 but can only get 0.4.8 to install via GitHub and CRAN. Any ideas about what might be happening?
  • Create New...