This repository contains R scripts used by the CCAMLR Secretariat to generate spatial layers, as well as examples of their use.
- Geospatial Rules
- Coastlines
- Polygons
- Candidate management units in Subarea 48.1
- Acoustic transects and stations in Subarea 48.1
The following rules, as endorsed by the Scientific Committee in 2023 (SC-CAMLR-42, paragraph 2.30), will be applied throughout. The rules will be updated if requested by the Scientific Committee.
-
geographical information system (GIS) objects use the EPSG 6932 projection (South Pole Lambert Azimuthal Equal-Area projection),
-
lines of more than 0.1 degree of longitude be densified,
-
polygon vertices be given clockwise in decimal degrees with at least five decimal places,
-
vertices be added where polygons meet (see WG-FSA-2023 Figure 1),
-
inland vertices be used for polygons that are bound by any coastline (continent and islands),
-
polygons be clipped to all coastlines (continent and islands) based on the most recent available coastline data,
-
the coastline be based on the latest available coastline data, as obtained from the SCAR Antarctic Digital Database (ADD) and other sources where needed (e.g., www.naturalearthdata.com),
-
analyses cite CCAMLR geospatial data (i.e., shapefiles) as CCAMLR. (Year). Geographical data layer: (Layer name). Version (Version), URL: (URL),
-
all maps cite data sources and projection used.
Starting in 2024, the CCAMLR Secretariat will maintain coastlines layers produced following the Geospatial Rules. Layers from two main data sources are used; from the UK Polar Data Centre (which generates the layers used in the SCAR ADD) and from Natural Earth (a public domain map dataset supported by the North American Cartographic Information Society).
More specifically, four datasets are used. These are:
- High resolution vector polygons of the Antarctic coastline
- Source: British Antarctic Survey / UK Polar Data Centre
- Short description: Coastline and ice shelves south of 60S
- URL: https://doi.org/10.5285/c7fe759d-e042-479a-9ecf-274255b4f0a1
- Citation: Gerrish, L., Ireland, L., Fretwell, P., & Cooper, P. (2023). High resolution vector polygons of the Antarctic coastline - VERSION 7.8 (Version 7.8) (Data set). NERC EDS UK Polar Data Centre. https://doi.org/10.5285/c7fe759d-e042-479a-9ecf-274255b4f0a1
- Vector polygons of the Sub-Antarctic coastline
- Source: British Antarctic Survey / UK Polar Data Centre
- Short description: Coastline between 50S and 60S (from which the data between 50W and 20W is extracted)
- URL: https://doi.org/10.5285/c1d83502-8799-4e3c-bdca-21db6a4405d4
- Citation: Gerrish, L. (2020). Vector polygons of the Sub-Antarctic coastline - VERSION 7.3 (Version 1.0) (Data set). UK Polar Data Centre, Natural Environment Research Council, UK Research & Innovation. https://doi.org/10.5285/c1d83502-8799-4e3c-bdca-21db6a4405d4
- Land polygons including major islands
- Source: Natural Earth
- Short description: 1:10m Physical Vectors for major land masses (from which the data between 40S and 60S is extracted)
- URL: https://www.naturalearthdata.com/downloads/10m-physical-vectors/
- Citation: Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com.
- Islands that are 2 sq. km or less in size
- Source: Natural Earth
- Short description: 1:10m Physical Vectors for minor islands (from which the data between 40S and 60S is extracted)
- URL: https://www.naturalearthdata.com/downloads/10m-physical-vectors/
- Citation: Made with Natural Earth. Free vector and raster map data @ naturalearthdata.com.
These data are combined, projected and simplified with a 10m tolerance using the CoastlineVx_x.R script. A final set of shapefiles is produced and available here (https://github.com/ccamlr/data/tree/main/geographical_data) and through the load_Coastline() function of the CCAMLRGIS R package (Version 4.0.7 and above). The following script shows how to plot the coastline while color-coding the data sources and types:
library(CCAMLRGIS)
#Load Coastline
Coast=load_Coastline()
#Plot
png(filename="Figures/Coastline.png",width=3000,height=3000,res=600)
par(mai=rep(0,4))
plot(st_geometry(Coast[Coast$source=="Natural Earth",]),col="orange",lwd=0.01)
plot(st_geometry(Coast[Coast$source=="BAS" & Coast$layer=="Land",]),col="blue",add=T,lwd=0.01)
plot(st_geometry(Coast[Coast$layer=="Ice shelves",]),col="grey",add=T,lwd=0.01)
plot(st_geometry(Coast[Coast$layer=="Ice tongues",]),col="green",add=T,lwd=0.01)
plot(st_geometry(Coast[Coast$layer=="Ice rumples",]),col="red",add=T,lwd=0.01)
legend("bottomleft",legend=c('Natural Earth land','BAS land','BAS ice shelves','BAS ice tongues','BAS ice rumples'),
fill=c('orange','blue','grey','green','red'),
seg.len=0,cex=0.75,
xpd=TRUE)
dev.off()
#> png
#> 2
Figure 2.1. CCAMLR coastline with elements color-coded by source. Sources: UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932.
The data contained in the shapefile is structured as follows (where version is the version of the CCAMLR coastline), with a row per set of polygons:
| version | source | srcvrsn | layer | surface |
|---|---|---|---|---|
| 1.0 | Natural Earth | Land V5.1.1 and Minor Islands V4.1.0 | Land | Land |
| 1.0 | BAS | Ant. coastline V7.8 and Sub-Ant. coastline V7.3 | Land | Land |
| 1.0 | BAS | Ant. coastline V7.8 | Ice shelves | Ice |
| 1.0 | BAS | Ant. coastline V7.8 | Ice tongues | Ice |
| 1.0 | BAS | Ant. coastline V7.8 | Ice rumples | Ice |
The following script shows how to plot a specific subset of the data (e.g., for Subarea 48.1), after rotating it so that North points up:
library(CCAMLRGIS)
#Load Coastline
Coast=load_Coastline()
#Load ASDs
ASDs=load_ASDs()
#Isolate Subarea 48.1
A481=ASDs[ASDs$GAR_Short_Label=="481",]
#Rotate objects
Lonzero=-60 #This longitude will point up
R_A481=Rotate_obj(A481,Lonzero)
R_coast=Rotate_obj(Coast,Lonzero)
#Create a bounding box for the region
bb=st_bbox(st_buffer(R_A481,10000)) #Get bounding box (x/y limits) + buffer
bx=st_as_sfc(bb) #Build spatial box to plot
#Use bounding box to crop coastline
R_coast=suppressWarnings( st_intersection(R_coast,bx) )
#Plot
png(filename='Figures/Coastline_481.png',width=2000,height=2400,res=300)
par(mai=rep(0.1,4)) #margins
plot(bx,col="lightblue")
plot(st_geometry(R_A481),border="black",lwd=2,add=T,col="palegreen")
plot(st_geometry(R_coast[R_coast$surface=="Land",]),col="grey",add=T)
plot(st_geometry(R_coast[R_coast$surface=="Ice",]),col="white",add=T)
add_RefGrid(bb=bb,ResLat = 2.5,ResLon = 5,lwd=1,fontsize = 0.75)
plot(bx,lwd=2,add=T,xpd=T)
legend(x=250000,y=2050000,
legend=c('Subarea 48.1','Ocean','Land', 'Ice Shelves'),
fill=c("palegreen","lightblue","grey","white"),
xpd=T)
dev.off()
#> png
#> 2
Figure 2.2. CCAMLR coastline for Subarea 48.1. Sources: UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932.
More examples can be found here.
This tutorial provides step-by-step instructions to build polygons while following the Geospatial Rules. Polygons may be used to represent areas (e.g., Research Blocks or Marine Protected Areas) or to subset data spatially (e.g. finding fishing locations that fall within a chosen area). Some of the operations used rely on the CCAMLRGIS R package which functions may be accessed and reviewed here.
Download the Blank Polygon Form, which is a .csv file with three columns (Name, Latitude and Longitude). While the name of these columns can be changed, these columns must be in that order. Fill the form, one row per vertex, giving coordinates with at least five decimal places, and clockwise. Repeat the polygon name for each of its vertices. As an example, here is a table of vertices for 3 polygons in Subarea 48.6 (the corresponding csv file is My_Polygons_Form.csv):
| Name | Latitude | Longitude |
|---|---|---|
| P1 | -65.00000 | -8.12345 |
| P1 | -65.00000 | 3.00000 |
| P1 | -72.00000 | 3.00000 |
| P1 | -72.00000 | -8.12345 |
| P2 | -63.00000 | 3.00000 |
| P2 | -63.00000 | 9.00000 |
| P2 | -68.00000 | 9.00000 |
| P2 | -68.00000 | 3.00000 |
| P3 | -61.00000 | 10.00000 |
| P3 | -61.00000 | 20.00000 |
| P3 | -65.54321 | 20.00000 |
| P3 | -65.54321 | 10.00000 |
Plotting these coordinates as they are, and connecting them with lines for each polygon, yields:
#Read the file
MyVertices=read.csv("Scripts/Polygons/My_Polygons_Form.csv")
#Plot
png(filename='Figures/MyPolygons0.png',width=2000,height=2000,res=300)
par(mai=c(0.9,0.9,0.2,0.2)) #margins
plot(MyVertices$Longitude,MyVertices$Latitude,xlab="Longitude",ylab="Latitude")
lines(MyVertices$Longitude[c(1:4,1)],MyVertices$Latitude[c(1:4,1)],col="red",lwd=3)
lines(MyVertices$Longitude[c(5:8,5)],MyVertices$Latitude[c(5:8,5)],col="green")
lines(MyVertices$Longitude[c(9:12,9)],MyVertices$Latitude[c(9:12,9)],col="blue")
text(-2.5,-68.5,"P1",col="red",cex=2)
text(6,-65.5,"P2",col="green",cex=2)
text(15,-63.2,"P3",col="blue",cex=2)
dev.off()
#> png
#> 2
Figure 3.1. Polygon coordinates as they are given in the ‘My_Polygons_Form.csv’ file.
As seen above, polygons P1 and P2 share a boundary. Following Geospatial Rule 4, vertices must be added where polygons meet as shown below:
Figure 3.2. Polygons A and B are each defined by four vertices and an additional vertex at the extremity of their shared edge (arrow). Figure taken from WG-FSA-2023 Fig. 1.
Vertices were added to the table, as indicated below:
| Name | Latitude | Longitude | |
|---|---|---|---|
| P1 | -65.00000 | -8.12345 | |
| P1 | -65.00000 | 3.00000 | |
| P1 | -68.00000 | 3.00000 | <= Added |
| P1 | -72.00000 | 3.00000 | |
| P1 | -72.00000 | -8.12345 | |
| P2 | -63.00000 | 3.00000 | |
| P2 | -63.00000 | 9.00000 | |
| P2 | -68.00000 | 9.00000 | |
| P2 | -68.00000 | 3.00000 | |
| P2 | -65.00000 | 3.00000 | <= Added |
| P3 | -61.00000 | 10.00000 | |
| P3 | -61.00000 | 20.00000 | |
| P3 | -65.54321 | 20.00000 | |
| P3 | -65.54321 | 10.00000 |
Now that the table of vertices is complete, the create_Polys() function of the CCAMLRGIS R package can be used to create densified and projected polygons:
library(CCAMLRGIS)
MyPolygons=create_Polys(MyVertices)
#Plot
png(filename='Figures/MyPolygons1.png',width=2000,height=2000,res=300)
par(mai=rep(0.1,4)) #margins
plot(st_geometry(MyPolygons)) #Plot all polygons to set axes limits
plot(st_geometry(MyPolygons[MyPolygons$ID=="P1",]),border="red",lwd=3,add=T)
plot(st_geometry(MyPolygons[MyPolygons$ID=="P2",]),border="green",add=T)
plot(st_geometry(MyPolygons[MyPolygons$ID=="P3",]),border="blue",add=T)
text(MyPolygons$Labx,MyPolygons$Laby,MyPolygons$ID,col=c("red","green","blue"),cex=2)
dev.off()
#> png
#> 2
Figure 3.3. Densified and projected polygons.
To clip polygons to the coastlines, use the coastline as generated in Section 2 above. Note that the coastline dataset includes ice shelves, while polygons must be clipped to the land:
#Load Coastline
Coast=load_Coastline()
#Isolate land and merge (union) polygons into one:
Land=Coast[Coast$surface=="Land",]
Land=st_union(Land)
#Clip polygons
MyPolygons=suppressWarnings( st_difference(MyPolygons,Land) )
#Plot
png(filename='Figures/MyPolygons2.png',width=2000,height=2000,res=300)
par(mai=rep(0.1,4)) #margins
plot(st_geometry(MyPolygons)) #Plot all polygons to set axes limits
plot(st_geometry(MyPolygons[MyPolygons$ID=="P1",]),border="red",lwd=3,add=T)
plot(st_geometry(MyPolygons[MyPolygons$ID=="P2",]),border="green",add=T)
plot(st_geometry(MyPolygons[MyPolygons$ID=="P3",]),border="blue",add=T)
text(MyPolygons$Labx,MyPolygons$Laby,MyPolygons$ID,col=c("red","green","blue"),cex=2)
dev.off()
#> png
#> 2
Figure 3.4. Polygons clipped to the coastline.
The spatial object that was created (MyPolygons) contains data in addition to the shapes of polygons:
| ID | AreaKm2 | Labx | Laby | geometry |
|---|---|---|---|---|
| P1 | 355139.0 | -107352.1 | 2399447 | POLYGON ((-390093.2 2740938… |
| P2 | 154794.3 | 284451.7 | 2706377 | POLYGON ((161468.4 2981429,… |
| P3 | 254156.2 | 765798.0 | 2857997 | POLYGON ((561536.3 3152447,… |
Each row corresponds to a polygon, the columns are:
-
ID: Polygon identifier, taken from the “Name” column in the table of vertices,
-
AreaKm2: Polygon area in square kilometers, calculated during creation,
-
Labx and Laby: Location of polygon centers, calculated during creation and used to label polygons,
-
geometry: POLYGON list of projected vertices (i.e., the shape and location of polygons)
Since polygons were clipped to the coastline after their creation, their areas and centers must be recalculated. Also, at this point, additional information may be added in the spatial object such as a reference to the paper describing these polygons:
#Update Areas
Ar=round(st_area(MyPolygons)/1000000,1)
MyPolygons$AreaKm2=as.numeric(Ar)
#Get labels locations
labs=st_coordinates(st_centroid(st_geometry(MyPolygons)))
MyPolygons$Labx=labs[,1]
MyPolygons$Laby=labs[,2]
#Add Reference
MyPolygons$Reference="WG-SAM-2023/xx Fig. z"| ID | AreaKm2 | Labx | Laby | geometry | Reference |
|---|---|---|---|---|---|
| P1 | 317949.4 | -109064.8 | 2439867 | POLYGON ((-390093.2 2740938… | WG-SAM-2023/xx Fig. z |
| P2 | 154794.3 | 284451.7 | 2706377 | POLYGON ((161468.4 2981429,… | WG-SAM-2023/xx Fig. z |
| P3 | 254156.2 | 765798.0 | 2857997 | POLYGON ((561536.3 3152447,… | WG-SAM-2023/xx Fig. z |
The polygons are now completed and may be exported:
#Export Shapefiles
st_write(MyPolygons,"Scripts/Polygons/Completed_Polygons.shp",append = F,quiet = T)Four files are generated, ending in “.dbf”, “.prj”, “.shp” and “.shx”. These can be zipped together, shared, and submitted along with the corresponding proposal.
The script below provides some elements to produce a map. Other examples are given here.
library(CCAMLRGIS)
library(terra)
library(png)
#Download bathymetry:
Bathy=load_Bathy(LocalFile=F,Res=5000) #Once downloaded, re-use it. See help(load_Bathy) for details
# Bathy=SmallBathy() #Use this instead for testing purposes first
#Load Coastline
Coast=load_Coastline()
#Load ASDs
ASDs=load_ASDs()
#Rotate objects
Lonzero=5 #This longitude will point up
R_bathy=Rotate_obj(Bathy,Lonzero)
R_asds=Rotate_obj(ASDs,Lonzero)
R_coast=Rotate_obj(Coast,Lonzero)
R_polys=Rotate_obj(MyPolygons,Lonzero)
#Update label location after rotation
labs=st_coordinates(st_centroid(st_geometry(R_polys)))
R_polys$Labx=labs[,1]
R_polys$Laby=labs[,2]
#Select ASD of interest
R_asdsb=R_asds[R_asds$GAR_Short_Label=="486",]
#Create a bounding box for the region
bb=st_bbox(st_buffer(R_asdsb,10000)) #Get bounding box (x/y limits) + buffer
bx=st_as_sfc(bb) #Build spatial box to plot
#Use bounding box to crop elements
R_asds=suppressWarnings(st_intersection(R_asds,bx))
R_coast=suppressWarnings(st_intersection(R_coast,bx))
R_bathy=crop(R_bathy,ext(bb))
#Plot
png(filename='Figures/MyPolygons3.png',width=2000,height=1350,res=300)
plot(R_bathy,breaks=Depth_cuts2,col=Depth_cols2,
legend=FALSE,axes=FALSE,mar=c(1,1.5,1,5.8),maxcell=5e6)
plot(st_geometry(R_asds),border="black",lwd=2,add=T)
plot(st_geometry(R_coast[R_coast$surface=="Land",]),col="grey",add=T)
plot(st_geometry(R_coast[R_coast$surface=="Ice",]),col="white",add=T)
plot(st_geometry(R_polys[R_polys$ID=="P1",]),border="red",add=T,col=rgb(1,0,0,alpha=0.3))
plot(st_geometry(R_polys[R_polys$ID=="P2",]),border="green",add=T,col=rgb(0,1,0,alpha=0.3))
plot(st_geometry(R_polys[R_polys$ID=="P3",]),border="blue",add=T,col=rgb(0,0,1,alpha=0.3))
text(R_polys$Labx,R_polys$Laby,R_polys$ID,col=c("red","green","blue"),cex=1.25)
text(0,3500000,"48.6",cex=1.5,font=2)
add_RefGrid(bb=bb,ResLat = 5,ResLon = 10,lwd=1,fontsize = 0.7)
plot(bx,lwd=2,add=T,xpd=T)
#Colorscale
add_Cscale(height=60,maxVal=-1,offset = 200,fontsize=0.65,width=15,lwd=1,
cuts = Depth_cuts2,
cols = Depth_cols2)
dev.off()
#> png
#> 2
Figure 3.5. Map of completed polygons. Sources: CCAMLR/UK Polar Data Centre/BAS/Natural Earth/GEBCO. Projection: EPSG 6932.
This section provides access to the resources used to update the candidate krill fishery management units and re-calculate their area when boundaries or coastlines are updated.
To date, two versions have been developed:
-
V1: as presented in Fig. 1 of SC-CAMLR-41. All files are in the KFMUs_V1 folder.
-
V2: as presented in Figs. 2–8 of WG-EMM-2024. All files are in the KFMUs_V2 folder.
Below are descriptions of the steps followed to build the latest version of shapefiles (V2). The Krill_Fishery_Management_Units_V2_0.R script follows three steps:
-
Step 1: Using a table of vertices, build polygons and clip them to the coastline,
-
Step 2: Isolate polygons, inspect plots (see Polygons_XX.png files), and remove problematic polygon parts (see Fig. 4.1 as an example),
-
Step 3: Update metadata (e.g., coastline version), export shapefiles (see EMM_24_Candidate_Krill_MUs_V2 files) and table of areas (see Table 4.1).
Once shapefiles are built and exported, the Plot_Krill_Fishery_Management_Units_V2.R script may be used to plot the corresponding map (see Fig. 4.2).
Figure 4.1. Example of problematic polygons (red) before (top) and after removal (bottom) from DP2 (left) and GS (right) candidate management units. Sources: CCAMLR/UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932.
Figure 4.2. Candidate management units after update. EI: Elephant Island, JOIN: Joinville, BS: Bransfield Strait, SSIW: South Shetland Islands West, GS: Gerlache Strait, DP: Drake Passage, PB: Powell Basin. Sources: CCAMLR/UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932.
Table 4.1. Updated candidate management units and their marine area.
| Candidate Management Unit | Marine Area (sq. km) |
|---|---|
| EI | 51669 |
| BS | 35208 |
| JOIN | 23033 |
| PB1 | 45456 |
| PB2 | 99236 |
| GS | 61088 |
| SSIW | 59293 |
| DP1 | 41688 |
| DP2 | 224045 |
This section provides access to the proposed acoustic transects and stations as developed by WG-ASAM-2024. All input files and a script to produce the maps below are located in this folder.
Figure 5.1. WG-ASAM-2024 proposed transects (red) and sampling stations (green) with a maximum spacing of 20 nautical miles. The proposed updated candidate management units (WG-ASAM-2024/11) are shown in black. Sources: GEBCO/CCAMLR/UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932 (rotated).
Figure 5.2. WG-ASAM-2024 proposed transects (red) and sampling stations (green) with a maximum spacing of 40 nautical miles. The proposed updated candidate management units (WG-ASAM-2024/11) are shown in black. Sources: GEBCO/CCAMLR/UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932 (rotated).
Figure 5.3. WG-ASAM-2024 proposed transects (red) and sampling stations (green) with a maximum spacing of 20 nautical miles within the Spatial Overlap Analysis footprint (SOA; WG-ASAM-2024/11) and of 40 nautical miles outside of the SOA footprint. The proposed updated candidate management units (WG-ASAM-2024/11) are shown in black. Sources: GEBCO/CCAMLR/UK Polar Data Centre/BAS and Natural Earth. Projection: EPSG 6932 (rotated).