18
Navigating through the R packages for movement Roc´ ıo Joo 1 , M.E. Boone 1 , T.A. Clay 2 , S.C. Patrick 2 , S. Clusella-Trullas 3 , M. Basille 1 1 University of Florida 2 University of Liverpool 3 Stellenbosch University useR! - July 2019 rocio.joo@ufl.edu rocio joo 1 / 15

Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

  • Upload
    others

  • View
    7

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Navigating through the R packages

for movement

Rocıo Joo1, M.E. Boone 1, T.A. Clay 2,S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille 1

1 University of Florida 2 University of Liverpool 3 Stellenbosch University

useR! - July 2019

Q [email protected] 7 rocio joo

1 / 15

Page 2: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Movement

Tracking packages: created to process or analyze tracking data

2 / 15

Page 3: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Movement

Tracking packages: created to process or analyze tracking data

2 / 15

Page 4: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Tracking workflow

3 / 15

Page 5: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Pre-processing

4 / 15

Page 6: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Pre-processing

E.g. Light level geolocator data

Packages: GeoLight, probGLS, FlightR, trackit, TripEstimation/SGAT, Twilight Free, telemetr,

animalTrack, TrackReconstruction

5 / 15

Page 7: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Pre-processing

E.g. Light level geolocator data

Packages: GeoLight, probGLS, FlightR, trackit, TripEstimation/SGAT, Twilight Free, telemetr,

animalTrack, TrackReconstruction

5 / 15

Page 8: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Post-processing

6 / 15

Page 9: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Post-processing

Cleaning (argosfilter, SDLfilter, T-LoCoH, TrajDataMining, trip)

Compression (adehabitatLT, amt, trajectories, trajr, TrajDataMining, rsMove)

7 / 15

Page 10: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Analysis

8 / 15

Page 11: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Path reconstruction

Purposes:

Correct errors

Finer data resolution

Regular time steps

Remember: model means assumptions!

Packages: HMMoce, kftrack, ukfsst/kfsst, ar-

gosTrack, bsam, crawl, ctmcmove, ctmm, Ba-

yesianAnimalTracker, TrackReconstruction

9 / 15

Page 12: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Behavioral pattern identification

Purpose:

Proxy of behaviors through theobserved movement patterns

Remember: model means assumptions!

Packages: EMbC, m2b, adehabitatLT, seg-

clust2d, bcpa, marcher, migrateR, bsam, ls-

mnsd, momentuHMM, moveHMM

10 / 15

Page 13: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Space use

Packages: adehabitatHR, adehabitatHS, amt, move, rhr, BBMM, ctmm, mkde, move,

movementAnalysis, T-LoCoH, hab, ctmcmove, feedr, VTrack, moveNT, recurse, rsMove

11 / 15

Page 14: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Quality of a package

Usability

Quality

Style

Documentation

how and why of eachfunction‘statement of need’:gap/improvement

12 / 15

Page 15: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Documentation

Survey: 225 users

11 pkgs over 75 % good/excellent documentation

adehabitatHR

adehabitatHSadehabitatLT

amtargosfilter

BBMM

bcpa bsam

crawl

ctmmEMbC

FLightR

GeoLight

hab

migrateR

momentuHMM

move

moveHMM

moveVis

recurse

rhr

SGAT/TripEstimation

T−LoCoH

trip

wildlifeDI

50

60

70

80

90

50 100 150

Number of respondents

Goo

d or

Exc

elle

nt R

atin

g

13 / 15

Page 16: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Documentation

Survey: 225 users 11 pkgs over 75 % good/excellent documentation

adehabitatHR

adehabitatHSadehabitatLT

amtargosfilter

BBMM

bcpa bsam

crawl

ctmmEMbC

FLightR

GeoLight

hab

migrateR

momentuHMM

move

moveHMM

moveVis

recurse

rhr

SGAT/TripEstimation

T−LoCoH

trip

wildlifeDI

50

60

70

80

90

50 100 150

Number of respondents

Goo

d or

Exc

elle

nt R

atin

g

13 / 15

Page 17: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Network

adehabitatHR

adehabitatHS

adehabitatLT

amt

animalTrack

anipaths

argosfilter

argosTrack

BayesianAnimalTracker

BBMM

bcpa

bsam

caribou

crawl

ctmcmove

ctmm

EMbC

feedr

FLightRGeoLight

hab

HMMocekftrack

lsmnsd

m2b

marcher

migrateR

mkde

momentuHMM

movemoveHMM

movementAnalysis

moveNT

moveVis

moveWindSpeed

probGLS

recurse

rhr

rpostgisLT

rsMove

SDLfilter

segclust2d

SGAT

SimilarityMeasures

SiMRiv

smam

T−LoCoH

telemetrtrackeR

trackit

TrackReconstruction

TrajDataMining

trajectories

trajr

trip

TwilightFree

ukfsst

VTrack

wildlifeDI

14 / 15

Page 18: Navigating through the R packages for movement · Navigating through the R packages for movement Roc o Joo1, M.E. Boone 1, T.A. Clay 2, S.C. Patrick 2, S. Clusella-Trullas 3, M. Basille

Thanks for your attention

location

(x, y, z, t)

Q [email protected] 7 rocio joo

15 / 15