inst/scripts/99.example.netmensuration.r
In AtlanticR/bio.groundfish: Core utilities to operate with groundfish data bases within Ecomod and associated projects.
# set up libs and bio functions
p = bio.groundfish::load.groundfish.environment( assessment.year=2016 )
# p = bio.groundfish::load.groundfish.environment( assessment.year=2016, netmensuration.years=2016 )
# steps required to recreate a local database of all data
recreate.perley.db = FALSE
if ( recreate.perley.db ) {
# define these in your Rprofile
# oracle.perley.user ="username"
# oracle.perley.password = "password"
# oracle.perley.db = "servername"
scanmar.db( DS="perley.datadump", p=p ) # ODBC data dump .. this step requires definition of password etc
scanmar.db( DS="perley.redo", p=p ) # perley had two db's merge them together: from XXXX-2002 and 2006 to 200X
## NOTE:: A cautionary note.. I am suspicious of the ID's that are given in the Perley data series ...
## Defualt matches using time/location suggest there might have been some errors when entered into the database..
}
# the following works upon many or annual time slices ( defined in p$netmensuration.years )
scanmar.db( DS="basedata.redo", p=p ) # Assimilate Scanmar files in raw data saves *.set.log files
scanmar.db( DS="basedata.lookuptable.redo", p=p ) # match modern data to GSINF positions and extract Mission/trip/set ,etc
scanmar.db( DS="sanity.checks.redo", p=p ) # QA/QC of data
# WARNING:: the following may crash as INLA does not exit gracefully from some errors
# and R cannot catch the faults .. restart R or reboot the system (it can happen)
# doing it year by year is probably wise for now
# usually insufficient data for these or just flat-lines .. no reliable data
if (FALSE) {
# challenging tests
totest = c("TEL2004529.20", "TEL2004529.16", "TEL2004530.84", "TEL2004530.85", "NED2015002.12",
"NED2010001.36", "NED2010002.1",
)
# nomatch in gsinf?
totest = c( "NED2010027.62", "NED2010027.117", "NED2010027.118", "NED2010027.201",
"NED2014101.16", "NED2014101.21", "NED2014101.14", "NED2014101.16", "NED2014101.21",
"NED2014002.1", "NED2014102.5",
"NED2014102.29", "NED2014102.33", "NED2014102.34", "NED2014102.37", "NED2014102.39", "NED2015002.57")
bc = scanmar.db( DS="bottom.contact.redo", p=p , debugid=totest ) ## a lot of data ???
# lots of 20 min tows in 2012 and 2014
# and some are really 15 min !
# to load a single result and view
# bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2010027.53") # strange wingspreads
# bottom.contact.plot( bc, netspread=TRUE )
}
scanmar.db( DS="bottom.contact.redo", p=p ) # bring in estimates of bottom contact times from scanmar
# swept areas are computed in bottom.contact.redo ..
# this step estimates swept area for those where there was insufficient data to compute SA directly from logs,
# estimate via approximation using speed etc.
groundfish.db( DS="sweptarea.redo", p=p )
figures.netmensuration( DS="all", p=p )
# netmind base data filtered for fishing periods .. not really used except for some plots
scanmar.db( DS="scanmar.filtered.redo", p=p )
create.marport.database = FALSE
if (create.marport.database ) {
p$netmensuration.years = 2013:2014
marport.db( DS="basedata.redo", p=p ) # load data from raw data files
marport.db( DS="gated.redo", p=p ) # QA/QC of data
marport = marport.db( DS="gated", p=p )
}
# --- misc stats / analysis
gs0 = scanmar.db( DS="bottom.contact", p=p ) # bring in estimates of bottom contact times from scanmar
gs0$timediff_official = as.numeric(gs0$edate - gs0$sdate) / 60 # that which would have been used (if at all .. )
gs0$bottom.dist = geosphere::distGeo( gs0[, c("bc.lon0","bc.lat0")], gs0[, c("bc.lon1", "bc.lat1")])/1000
# settype: 1=stratified random, 2=regular survey, 3=unrepresentative(net damage),
# 4=representative sp recorded(but only part of total catch), 5=comparative fishing experiment,
# 6=tagging, 7=mesh/gear studies, 8=explorartory fishing, 9=hydrography
# setype
table(gs0$settype)
1 3 4 5 8 9
5333 294 1 608 147 59
# settype vs mean duration
tapply( gs0$bottom_duration/60, gs0$settype, mean, na.rm=TRUE )
1 3 4 5 8 9
29.76567 NaN NaN 28.45052 NaN NaN
w2a = which( gs0$geardesc == "Western IIA trawl" & gs0$settype %in% c(1,2,5) ) # for distribution checks for western IIA trawl
gs = gs0[ w2a, ]
# trawl duration bias
plot( jitter(gs$timediff_official), jitter(gs$bottom_duration/60), xlim=c(10,40), cex=0.65, col="slateblue",
xlab="Trawl duration: official (minutes)", ylab="Trawl duration: computed (minutes)" )
abline(a=0,b=1, col="grey" )
# distance bias
plot( jitter(gs$dist_km), jitter(gs$bottom.dist), xlim=c(1.5,4),ylim=c(0,5),cex=0.65, col="slateblue",
xlab="Trawl length: official (km)", ylab="Trawl length: computed from end points (km)" )
abline(a=0,b=1, col="grey" )
# distance bias
plot( jitter(gs$bc.dist), jitter(gs$bottom.dist), xlim=c(1.5,6),ylim=c(0,6),cex=0.65, col="slateblue",
xlab="Trawl length: official (km)", ylab="Trawl length: computed from track(km)" )
abline(a=0,b=1, col="grey" )
# distance bias ( vert has no influence)
plot( jitter(gs$bc.dist), jitter(gs$bc.dist.h), xlim=c(1.5,4),ylim=c(0,5),cex=0.65, col="slateblue",
xlab="Trawl length: total (km)", ylab="Trawl length: horiz (km)" )
abline(a=0,b=1, col="grey" )
# swept area (door width X distance)
plot( door.sa ~ as.factor(yr), gs[ gs$yr> 2009,], ylab="Door width (m)", xlab="Year" )
plot( door.sa ~ log(bottom_depth), gs[ ,], xlim=log(c(20,650)) )
> plot( door.sa ~ log(bottom_depth), gs[ ,], xlim=log(c(20,650)) )
>
# Swept area vs depth
plot( (door.sa) ~ log(bc.depth.mean), gs[ ,], col="slategray", cex=0.5, xlab="log Depth (m)", ylab="Swept area (km^2)" )
# points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2010 ,], pch="*", col="red" )
points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2011 ,], pch=20, col="steelblue", cex=1.25 )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2012 ,], pch=22, col="orange" )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2013 ,], pch=19, col="brown", cex= 0.8 )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2014 ,], pch=19, col="brown", cex= 0.8 )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2015 ,], pch=19, col="red", cex= 0.8 )
plot( bottom_duration ~ as.factor( strat), gs[ gs$yr> 2008,] )
plot( bottom_duration ~ bottom_salinity, gs[ ,], xlim=c(31,36) )
plot( door.sa ~ bottom_salinity, gs[ ,], xlim=c(31,36) )
plot( door.sa ~ bottom_depth, gs[ ,], xlim=c(31,36) )
plot( door.sa ~ log(bottom_depth), gs[ ,], xlim=log(c(20,650)) )
plot( bottom_duration ~ as.factor(yr), gs[ gs$yr> 2008,] )
plot( (bottom_duration) ~ bc.depth.mean, gs[ ,], pch=".", col="red" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2010 ,], pch="*", col="red" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2011 ,], pch="*", col="blue" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2012 ,], pch=22, col="orange" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2013 ,], pch=19, col="brown", cex= 0.8 )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2014 ,], pch=19, col="brown", cex= 0.8 )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2015 ,], pch=19, col="red", cex= 0.8 )
pp = tapply( gs$id, year(gs$bc0.datetime), function(x) { length(unique(x))} )
pp = data.frame( yr= rownames(pp), n.bc=pp)
#scanmar.db( DS="bottom.contact.redo", p=p ) # bring in estimates of bottom contact times from scanmar
oo = tapply( gs$id, year(gs$sdate), function(x) { length(unique(x))} )
oo = data.frame( yr = rownames(oo), n.gs= oo)
length( gs[ which(is.finite(gs$bottom_duration) ), "id" ] )
length( gs[ which(is.finite(gs$bottom_duration) & gs$bc0.sd<30 & gs$bc1.sd<30), "id" ] )
sc = scanmar.db( DS="sanity.checks", p=p ) # QA/QC of data
rr = tapply( sc$id, sc$year, function(x) { length(unique(x))} )
rr = data.frame( yr= rownames(rr), n.scanmar=rr )
nm = scanmar.db( DS="scanmar.filtered", p=p ) # bring in estimates of bottom contact times from scanmar
qq = tapply( nm$id, nm$year, function(x) { length(unique(x))} )
qq = data.frame( yr= rownames(qq), n.filtered=qq )
res = merge ( oo, pp, by="yr")
res = merge ( res, qq, by="yr")
res = merge ( res, rr, by="yr")
# ----
# debugging SA estimates
g = scanmar.db( DS="bottom.contact", p=p )
plot( dist_km~ I(wing.sa / wing.mean*1000), g, ylim=c(0,5) )
gr = abs(g$dist_km - (g$wing.sa / g$wing.mean)*1000)
strange = which ( gr > 1 & g$gear==9 & g$settype==1 )
g[strange, "id"]
"NED2014101.23" "NED2011002.6" "NED2014101.15" "NED2015002.20"
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2014018.71") # depth sensor not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2013022.192") # large depth range
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2013022.205") # depth sensor not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2013022.208") # GPS not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2011002.53") #
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2011002.45") # doorspread failure and almost no wingspread
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2014101.23")
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2011002.6")
bottom.contact.plot( bc, netspread=TRUE )
bc = scanmar.db( DS="bottom.contact.redo", p=p , debugid= "NED2011002.45") # GPS not working
bc = scanmar.db( DS="bottom.contact.redo", p=p , debugid= "NED2013022.208") # GPS not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2009002.17") # doorspread failure and almost no wingspread
bottom.contact.plot( bc, netspread=TRUE )
plot( longitude ~ latitude, bc$plotdata )
plot( wingspread ~ ts, bc$plotdata )
plot( doorspread ~ ts, bc$plotdata )
uu = scanmar.db( DS="sanity.checks", p=p, YR=2011 )
uu = scanmar.db( DS="basedata", p=p, YR=2011 ) # Assimilate Scanmar files in raw data saves *.set.log files
vv = scanmar.db( DS="basedata.lookuptable", p=p )
ww = which( vv$id=="NED2009027.25" )
xx = uu[ which( uu$nm_id == vv$nm_id[ww] ) , ]
plot( wingspread ~ timestamp, xx )
plot( doorspread ~ timestamp, xx )
AtlanticR/bio.groundfish documentation built on May 28, 2019, 11:05 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# set up libs and bio functions
p = bio.groundfish::load.groundfish.environment( assessment.year=2016 )
# p = bio.groundfish::load.groundfish.environment( assessment.year=2016, netmensuration.years=2016 )
# steps required to recreate a local database of all data
recreate.perley.db = FALSE
if ( recreate.perley.db ) {
# define these in your Rprofile
# oracle.perley.user ="username"
# oracle.perley.password = "password"
# oracle.perley.db = "servername"
scanmar.db( DS="perley.datadump", p=p ) # ODBC data dump .. this step requires definition of password etc
scanmar.db( DS="perley.redo", p=p ) # perley had two db's merge them together: from XXXX-2002 and 2006 to 200X
## NOTE:: A cautionary note.. I am suspicious of the ID's that are given in the Perley data series ...
## Defualt matches using time/location suggest there might have been some errors when entered into the database..
}
# the following works upon many or annual time slices ( defined in p$netmensuration.years )
scanmar.db( DS="basedata.redo", p=p ) # Assimilate Scanmar files in raw data saves *.set.log files
scanmar.db( DS="basedata.lookuptable.redo", p=p ) # match modern data to GSINF positions and extract Mission/trip/set ,etc
scanmar.db( DS="sanity.checks.redo", p=p ) # QA/QC of data
# WARNING:: the following may crash as INLA does not exit gracefully from some errors
# and R cannot catch the faults .. restart R or reboot the system (it can happen)
# doing it year by year is probably wise for now
# usually insufficient data for these or just flat-lines .. no reliable data
if (FALSE) {
# challenging tests
totest = c("TEL2004529.20", "TEL2004529.16", "TEL2004530.84", "TEL2004530.85", "NED2015002.12",
"NED2010001.36", "NED2010002.1",
)
# nomatch in gsinf?
totest = c( "NED2010027.62", "NED2010027.117", "NED2010027.118", "NED2010027.201",
"NED2014101.16", "NED2014101.21", "NED2014101.14", "NED2014101.16", "NED2014101.21",
"NED2014002.1", "NED2014102.5",
"NED2014102.29", "NED2014102.33", "NED2014102.34", "NED2014102.37", "NED2014102.39", "NED2015002.57")
bc = scanmar.db( DS="bottom.contact.redo", p=p , debugid=totest ) ## a lot of data ???
# lots of 20 min tows in 2012 and 2014
# and some are really 15 min !
# to load a single result and view
# bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2010027.53") # strange wingspreads
# bottom.contact.plot( bc, netspread=TRUE )
}
scanmar.db( DS="bottom.contact.redo", p=p ) # bring in estimates of bottom contact times from scanmar
# swept areas are computed in bottom.contact.redo ..
# this step estimates swept area for those where there was insufficient data to compute SA directly from logs,
# estimate via approximation using speed etc.
groundfish.db( DS="sweptarea.redo", p=p )
figures.netmensuration( DS="all", p=p )
# netmind base data filtered for fishing periods .. not really used except for some plots
scanmar.db( DS="scanmar.filtered.redo", p=p )
create.marport.database = FALSE
if (create.marport.database ) {
p$netmensuration.years = 2013:2014
marport.db( DS="basedata.redo", p=p ) # load data from raw data files
marport.db( DS="gated.redo", p=p ) # QA/QC of data
marport = marport.db( DS="gated", p=p )
}
# --- misc stats / analysis
gs0 = scanmar.db( DS="bottom.contact", p=p ) # bring in estimates of bottom contact times from scanmar
gs0$timediff_official = as.numeric(gs0$edate - gs0$sdate) / 60 # that which would have been used (if at all .. )
gs0$bottom.dist = geosphere::distGeo( gs0[, c("bc.lon0","bc.lat0")], gs0[, c("bc.lon1", "bc.lat1")])/1000
# settype: 1=stratified random, 2=regular survey, 3=unrepresentative(net damage),
# 4=representative sp recorded(but only part of total catch), 5=comparative fishing experiment,
# 6=tagging, 7=mesh/gear studies, 8=explorartory fishing, 9=hydrography
# setype
table(gs0$settype)
1 3 4 5 8 9
5333 294 1 608 147 59
# settype vs mean duration
tapply( gs0$bottom_duration/60, gs0$settype, mean, na.rm=TRUE )
1 3 4 5 8 9
29.76567 NaN NaN 28.45052 NaN NaN
w2a = which( gs0$geardesc == "Western IIA trawl" & gs0$settype %in% c(1,2,5) ) # for distribution checks for western IIA trawl
gs = gs0[ w2a, ]
# trawl duration bias
plot( jitter(gs$timediff_official), jitter(gs$bottom_duration/60), xlim=c(10,40), cex=0.65, col="slateblue",
xlab="Trawl duration: official (minutes)", ylab="Trawl duration: computed (minutes)" )
abline(a=0,b=1, col="grey" )
# distance bias
plot( jitter(gs$dist_km), jitter(gs$bottom.dist), xlim=c(1.5,4),ylim=c(0,5),cex=0.65, col="slateblue",
xlab="Trawl length: official (km)", ylab="Trawl length: computed from end points (km)" )
abline(a=0,b=1, col="grey" )
# distance bias
plot( jitter(gs$bc.dist), jitter(gs$bottom.dist), xlim=c(1.5,6),ylim=c(0,6),cex=0.65, col="slateblue",
xlab="Trawl length: official (km)", ylab="Trawl length: computed from track(km)" )
abline(a=0,b=1, col="grey" )
# distance bias ( vert has no influence)
plot( jitter(gs$bc.dist), jitter(gs$bc.dist.h), xlim=c(1.5,4),ylim=c(0,5),cex=0.65, col="slateblue",
xlab="Trawl length: total (km)", ylab="Trawl length: horiz (km)" )
abline(a=0,b=1, col="grey" )
# swept area (door width X distance)
plot( door.sa ~ as.factor(yr), gs[ gs$yr> 2009,], ylab="Door width (m)", xlab="Year" )
plot( door.sa ~ log(bottom_depth), gs[ ,], xlim=log(c(20,650)) )
> plot( door.sa ~ log(bottom_depth), gs[ ,], xlim=log(c(20,650)) )
>
# Swept area vs depth
plot( (door.sa) ~ log(bc.depth.mean), gs[ ,], col="slategray", cex=0.5, xlab="log Depth (m)", ylab="Swept area (km^2)" )
# points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2010 ,], pch="*", col="red" )
points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2011 ,], pch=20, col="steelblue", cex=1.25 )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2012 ,], pch=22, col="orange" )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2013 ,], pch=19, col="brown", cex= 0.8 )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2014 ,], pch=19, col="brown", cex= 0.8 )
#points( (door.sa) ~ log( bc.depth.mean), gs[gs$yr==2015 ,], pch=19, col="red", cex= 0.8 )
plot( bottom_duration ~ as.factor( strat), gs[ gs$yr> 2008,] )
plot( bottom_duration ~ bottom_salinity, gs[ ,], xlim=c(31,36) )
plot( door.sa ~ bottom_salinity, gs[ ,], xlim=c(31,36) )
plot( door.sa ~ bottom_depth, gs[ ,], xlim=c(31,36) )
plot( door.sa ~ log(bottom_depth), gs[ ,], xlim=log(c(20,650)) )
plot( bottom_duration ~ as.factor(yr), gs[ gs$yr> 2008,] )
plot( (bottom_duration) ~ bc.depth.mean, gs[ ,], pch=".", col="red" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2010 ,], pch="*", col="red" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2011 ,], pch="*", col="blue" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2012 ,], pch=22, col="orange" )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2013 ,], pch=19, col="brown", cex= 0.8 )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2014 ,], pch=19, col="brown", cex= 0.8 )
points( (bottom_duration) ~ bc.depth.mean, gs[gs$yr==2015 ,], pch=19, col="red", cex= 0.8 )
pp = tapply( gs$id, year(gs$bc0.datetime), function(x) { length(unique(x))} )
pp = data.frame( yr= rownames(pp), n.bc=pp)
#scanmar.db( DS="bottom.contact.redo", p=p ) # bring in estimates of bottom contact times from scanmar
oo = tapply( gs$id, year(gs$sdate), function(x) { length(unique(x))} )
oo = data.frame( yr = rownames(oo), n.gs= oo)
length( gs[ which(is.finite(gs$bottom_duration) ), "id" ] )
length( gs[ which(is.finite(gs$bottom_duration) & gs$bc0.sd<30 & gs$bc1.sd<30), "id" ] )
sc = scanmar.db( DS="sanity.checks", p=p ) # QA/QC of data
rr = tapply( sc$id, sc$year, function(x) { length(unique(x))} )
rr = data.frame( yr= rownames(rr), n.scanmar=rr )
nm = scanmar.db( DS="scanmar.filtered", p=p ) # bring in estimates of bottom contact times from scanmar
qq = tapply( nm$id, nm$year, function(x) { length(unique(x))} )
qq = data.frame( yr= rownames(qq), n.filtered=qq )
res = merge ( oo, pp, by="yr")
res = merge ( res, qq, by="yr")
res = merge ( res, rr, by="yr")
# ----
# debugging SA estimates
g = scanmar.db( DS="bottom.contact", p=p )
plot( dist_km~ I(wing.sa / wing.mean*1000), g, ylim=c(0,5) )
gr = abs(g$dist_km - (g$wing.sa / g$wing.mean)*1000)
strange = which ( gr > 1 & g$gear==9 & g$settype==1 )
g[strange, "id"]
"NED2014101.23" "NED2011002.6" "NED2014101.15" "NED2015002.20"
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2014018.71") # depth sensor not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2013022.192") # large depth range
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2013022.205") # depth sensor not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2013022.208") # GPS not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2011002.53") #
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2011002.45") # doorspread failure and almost no wingspread
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2014101.23")
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2011002.6")
bottom.contact.plot( bc, netspread=TRUE )
bc = scanmar.db( DS="bottom.contact.redo", p=p , debugid= "NED2011002.45") # GPS not working
bc = scanmar.db( DS="bottom.contact.redo", p=p , debugid= "NED2013022.208") # GPS not working
bc = scanmar.db( DS="bottom.contact", p=p , setid= "NED2009002.17") # doorspread failure and almost no wingspread
bottom.contact.plot( bc, netspread=TRUE )
plot( longitude ~ latitude, bc$plotdata )
plot( wingspread ~ ts, bc$plotdata )
plot( doorspread ~ ts, bc$plotdata )
uu = scanmar.db( DS="sanity.checks", p=p, YR=2011 )
uu = scanmar.db( DS="basedata", p=p, YR=2011 ) # Assimilate Scanmar files in raw data saves *.set.log files
vv = scanmar.db( DS="basedata.lookuptable", p=p )
ww = which( vv$id=="NED2009027.25" )
xx = uu[ which( uu$nm_id == vv$nm_id[ww] ) , ]
plot( wingspread ~ timestamp, xx )
plot( doorspread ~ timestamp, xx )
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.