Data for: "How flexible are habitat specialists? Short-term space use in obligate coral-dwelling damselfishes"

Full description

This dataset contains the data associated with the following publication: 
Streit RP, Hemingson CR, Cumming GS, Bellwood DR (2021) How flexible are habitat specialists? Short-term space use in obligate coral-dwelling damselfishes. Reviews in Fish Biology and Fisheries

In this study, I visually tracked damselfishes (Pomacentrus moluccensis and Chromis viridis) underwater across 5 sites in the lagoon of Lizard Island. 
Tracking data was recorded daily over six days for each fish, marking locations every 15 seconds for five minutes per fish, per day.
Locations were drawn on "mud maps" on underwater paper and later transferred to digital orthomosaic files created with Agisoft Photoscan.
These maps, as well as fish locations, were later geo-referenced in ArcGIS.
From georeferenced location coordinates, I calculated MCPs (minimum convex polygons) as well as KUDs (kernel utilisation distributions) for each fish for each day and one cumulative area measure across all observation days.
Later I assessed habitat quality (proportional coral cover) and complexity metrics within each fish usage area. Using DEMs (digital elevation models) from each study site, resolution 1cm,
I also measured sitewide habitat quality (coral cover and rugosity) by creating 100 random points in ArcGIS on each site and measuring proportional coral cover and rugosity within an average sized (average MCP across all fish) circle that was placed centrally on each randomly generated point. 
To compare the complexity measures to something relatable I also calculated complexity measures of "typical reef habitat", i.e. sand, rubble, massive coral and branching coral.
To do this I placed circles with the average MCP area (across all fish) size over the respective habitats, using my photomosaic maps in ArcGIS.

This data record contains three data files with slightly different data: 

A) "Raw Fish Location Coordinates"
(this dataset contains the GPS coordinates for each manually tracked location across all fishes and all observation days. Georeferenced to projection: WGS 84/UTM 55S)

SITE:    ID of study sites, 1 to 4 as well as site5 and site5MOSAIC, Site5MOSAIC contains Site5A, but stretches further along the reef 
TAG:    Individual fish tag code (based on two lateral coloured lines of visual implant elastomer), three letters describe colour of top line, colour of bottom line and side of body (B = blue, R = red, G = green, X = black)
OBSERVER:    observer of the fish, BERT = Robert Streit, CHRIS = Christopher Hemingson
DAY:    Consecutive observation days, starting one day after tagging
SPECIES:    species ID, either Pomacentrus mollucensis or Chromis viridis
x:    x-coordinate of GPS point
y:    y-coordinate of GPS point

B) "Area measures" 
(This dataset contains the processed data, based on above location coordinates, showing measured area sizes per fish and habitat quality metrics for each area)

SITE:    ID of study sites, 1 to 4 as well as site5 and site5MOSAIC, Site5MOSAIC contains Site5A, but stretches further along the reef 
SPECIES:    species ID, either Pomacentrus mollucensis or Chromis viridis
OBSERVER:    observer of the fish, BERT = Robert Streit, CHRIS = Christopher Hemingson
TAG:    Individual fish tag code (based on two lateral coloured lines of visual implant elastomer), three letters describe colour of top line, colour of bottom line and side of body (B = blue, R = red, G = green, X = black)
TL:    Total length of the fish in mm (measured using callipers during tagging)
AREA_TYPE:    Type of area measure, either based on minimum convex polygons (MCP) or kernel Utilisation distributions (KUD), core areas (KUD_50) and extents (KUD_95) are distinguished
WEEKLY.AREA:    Cumulative estimate over all observation days, i.e. all locations of one fish pooled together. 
MEAN.Stability:    "Stability" should rather be "consistency" and is a measure of how consistenly each fish used its area. Calculated as the proportion of Area on Day_n+1 that lies within the area of Day_n, this was calculated for all consecutive day pairs, and the averaged across the week to have one value for each fish
SD.Stability:    Standard deviation, rather than mean, of the above measure, i.e. across all consecutive day pairs for each fish
SEM.Stability:    Standard error of the mean of the above measure
MIN.Stability:    Minimum of the above measure
MAX.Stability:    Maximum of the above measure
n_DAYS:    Number of days each fish was under observation.
LIVECORAL.WEEKLY.UD50:    Proportional live coral within each fish's weekly (all days pooled) core area (KUD_50), produced by placing 100 random points over the core area in ArcGIS and visually assessing live coral
Elevation.MEAN:    Average Elevation within each weekly core area, calculated from DEM (resolution 1cm), i.e. average across all pixels contained within area. 
Elevation.sd:    Standard deviation of elevation
Elevation.SEM:    Standard error of the mean of elevation
Slope.MEAN:    Average Slope within each weekly core area, calculated from DEM (resolution 1cm), i.e. average across all pixels contained within area. (calculated with terrain function in R)
Slope.sd    standard deviation of slope
Slope.SEM:    standard error of the mean of slope
Roughness.MEAN:    Average Roughness within each weekly core area, calculated from DEM (resolution 1cm), i.e. average across all pixels contained within area. (calculated with terrain function in R)
Roughness.sd:    standard deviation of Roughness
Roughness.SEM:    standard error of the mean of Roughness
RugosityRatio:    Rugosity Ratio of the weekly core area, calculated from the following two values (SurfaceArea / Planar Area)
SurfaceArea:    Surface Area of the weekly core area, accounting for structure, calculted in R
PlanarArea:    Planar area of the weekly core area, calculated as plain area measure (i.e. how many pixels within area / divided by DEM resolution)
SITE.MEAN.Rugosity:    Mean rugosity on a site level, calculate by placing 100 random point across sites and placing a circle of the mean MCP area on top of each point, then calculating rugosity within each circle and taking an average of the 100 circles, i.e. estimating potential 'randomly placed damselfish areas'
SITE.SEM.Rugosity:    Standard Error of site-level rugosity
SITE.LIVECORAL:    Site level live coral, calculated by creating 100 random points in ArcGIS across all sites and visually identifying live coral

C) "Real World Complexity Measures"
(This dataset contains "habitat quality measures" for "real world" examples, i.e. visually chosen locations (using photomoaic maps in ArcGIS) at each site that represent "typical" habitat conditions.)

SITE:    ID of study sites, 1 to 4 as well as site5 and site5MOSAIC, Site5MOSAIC contains Site5A, but stretches further along the reef 
BENTHOS:    Type of benthic structure selected, i.e. Sand, Rubble, Branching Coral and Live Coral. A circle of avergae MCP area sizes across all fishes was placed over the corresponding habitat at each site
REALWORLD_LIVECORAL:    Livecoral proportion, based on 100 randomly generated points within each circle in ArcGIS
REALWORLD_RugosityRatio:    Rugosity Ratio of the circles, calculated using DEMs and R as above. 

Software/equipment used to create/collect the data: Visual underwater surveys, ArcGIS, Agisoft Metashape Structure from Motion Software

Software/equipment used to manipulate/analyse the data: R and R-Studio, ArcGIS.