R/buildAstroStats.r
In tmcd82070/CAMP_RST: CAMP RST R Routines
Defines functions
buildAstroStats
Documented in
buildAstroStats
#' @export
#'
#' @title buildAstroStats
#'
#' @description Given a particular trap sequence, calculate the proportion of
#' time spent fishing during nighttime and while the moon was up.
#'
#' @param release.visit A data frame containing release data obtained from the
#' release-data SQL sequence.
#'
#' @param visit.df A data frame containing visits obtained from the catch-data
#' SQL sequence.
#'
#' @return A dataframe with one row for all unique \code{batchDates} included
#' within \code{min.date} and \code{max.date}, inclusive. All variables, as
#' provided via \code{varVec}, have a non-\code{NA} value.
#'
#' @details The \code{buildAstronStats} function first queries the CAMP database
#' for all relevant visits via the \code{Visits} table, as well as the
#' astronomical sun and moon rise and set times, contained within table
#' \code{Dates}. The "Not Fishing" data are also obtained, via a query to
#' table \code{TempSumUnmarkedByTrap_Run_Final} in the underlying CAMP
#' \code{mdb}. Because of this, the catch SQL sequence must be run prior to
#' the same of the efficiency.
#'
#' The function emulates the calculation of \code{SampleMinutes} performed by
#' the catch SQL sequence. This is done in order to calculate denominators
#' for calculation of the proportion of time spent fishing at night and while
#' the moon is up. The \code{SampleMinutes} field from
#' \code{TempSumUnmarkedByTrap_Run_Final} was not utilized because the
#' astronomical metrics need to calculate fishing with respect to the included
#' fishing instances in the calculation of efficiency. These include fishing
#' instances regardless of the value of the \code{includeCatchID} field. The
#' value of \code{SampleMinutes} in the \code{TempSumUnmarkedByTrap_Run_Final}
#' calculated for catch estimation excludes bad fishing, as identified via the
#' \code{includecatchID} field. Therefore, re-estimation of
#' \code{SampleMinutes}, based on efficiency concerns, must be performed.
#' These \code{SampleMinutes} are called \code{JasonSampleMinutes}, so as to
#' differentiate. Values of this variable equals \code{-99} for the first
#' visit of a trapping sequence, and \code{-88} in the case of a time frame
#' greater than the gap in fishing length. Currently, the fishing-gap length
#' is set via global variable \code{fishingGapMinutes}, and equals
#' \code{10080} minutes.
#'
#' Calculation of astronomical proportions utilized \code{campR} functions
#' \code{makeSkinnyTimes} and \code{getTimeProp}.
#'
#' @seealso \code{makeSkinnyTimes}, \code{getTimeProp}
#'
#' @author WEST Inc.
#'
#' @examples
#' \dontrun{
#' df <- buildAstroStats(release.visit=release.visit,
#' visit.df=visit.df)
#' }
buildAstroStats <- function(release.visit,visit.df){
# release.visit <- release.visit
# visit.df <- visit.df
# ----- Define some convenient dates.
min.date2 <- "1990-01-01"
max.date2 <- Sys.Date()
# ---- Grab the time zone.
time.zone <- get( "time.zone", envir=.GlobalEnv )
fishingGapMinutes <- get("fishingGapMinutes",envir=.GlobalEnv)
# ---- Open ODBC channel.
db <- get( "db.file", envir=.GlobalEnv )
ch <- odbcConnectAccess(db)
# max.date.eff <- max(release.visit$ReleaseDate) + 30*24*60*60 # 30-day buffer after
# min.date.eff <- min(release.visit$ReleaseDate) - 30*24*60*60 # 30-day buffer before
max.date.eff <- as.POSIXct(max.date2,format="%Y-%m-%d",tz="UTC") + 30*24*60*60
min.date.eff <- as.POSIXct(min.date2,format="%Y-%m-%d",tz="UTC") - 30*24*60*60
# ---- Get dates information for moon and sun info.
tblDates <- sqlQuery(ch,paste0("SELECT uniqueDate,
nightLength,
moonRise,
moonSet,
sunRise,
sunSet
FROM Dates
WHERE uniqueDate <= #",as.Date(max.date.eff),"#
AND uniqueDate >= #",as.Date(min.date.eff),"#
ORDER BY uniqueDate"))
trapVisits <- sqlQuery(ch,paste0("SELECT trapVisitID,
trapPositionID,
visitTime,
visitTime2,
visitTypeID
FROM trapVisit
WHERE visitTime <= #",as.Date(max.date2),"#
AND visitTime >= #",as.Date(min.date2),"#
AND ( (visitTypeID < 5 AND fishProcessedID <> 2)
OR (visitTypeID = 1 AND fishProcessedID = 2) )
ORDER BY trapPositionID,visitTime"))
# ---- Need to make sure this table exists first, when this code is finalized.
notFishing <- sqlQuery(ch,paste0("SELECT SampleDate,
StartTime,
EndTime,
TrapPositionID AS oldtrapPositionID,
SampleMinutes
FROM TempSumUnmarkedByTrap_Run_Final
WHERE TrapStatus = 'Not Fishing'
ORDER BY TrapPositionID, EndTime"))
close(ch)
# ---- We compile metrics that we need to compile over trapping.
# ---- Construct a trapVisit data frame of all visits.
tmp <- trapVisits
#tmp$fishProcessedID <- NULL
tmp$visitTime <- as.POSIXlt(strftime(trapVisits$visitTime),tz=time.zone)
tmp$visitTime2 <- as.POSIXlt(strftime(trapVisits$visitTime2),tz=time.zone)
# ---- Bring in the mean forkLengths.
fl <- attr(visit.df,"fl")
# ---- Some trapVisitIDs are not in the final catch table, for whatever reason. Find these
# ---- missing trap instances so they can be zeroed out. These trap visits often are found
# ---- in dfs when calculating efficiency. We will put their fish at n of fish at 0.
maxTrapVisitID <- max(trapVisits$trapVisitID)
trapVisitIDSpine <- data.frame(trapVisitID=seq(1,maxTrapVisitID,1))
fl0 <- merge(trapVisitIDSpine,fl,by=c("trapVisitID"),all.x=TRUE)
fl0[is.na(fl0)] <- 0
fl0 <- fl0[fl0$nForkLength == 0,]
tmp <- merge(tmp,fl,by=c("trapVisitID"),all.x=TRUE)
# ---- Construct start and end times. I also construct SampleMinutes...sometimes these differ from Connie's, by 60
# ---- minutes. This has to do with daylight savings. I need POSIX to 'be dumb' with respect to daylight savings,
# ---- to match Connie. I wonder if Connie's SampleMinutes are off by 60 minutes? I suspect the times recorded in
# ---- the CAMP are 'raw' times, and so when one "springs forward" +60 minutes go along for the ride, and when one
# ---- "falls back," CAMP loses 60 minutes. This explains why spring-time SampleMinutes that I calculate are short
# ---- by 60 minutes. I don't think this is easily fixed.
tmp <- tmp[order(tmp$trapPositionID,tmp$visitTime),]
# ---- Apply the lag throughout.
tmp$StartTime <- as.POSIXlt(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone)
# ---- Identify where the lag logic is wrong. These occur when the lag visitTime != lag visitTime2. Adjust the StartTime to be correct.
tmp$StartTime <- ifelse(as.POSIXlt(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone) !=
as.POSIXlt(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime2[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone),
as.POSIXct(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime2[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone),
as.POSIXct(tmp$StartTime))
# ---- So POSIX is awful with ifelse. Or I'm not doing it right. Regardless, put it to how we want it.
tmp$StartTime2 <- as.POSIXlt(tmp$StartTime,format="%Y-%m-%d %H:%M:%S",tz=time.zone,origin="1970-01-01 00:00:00 UTC")
tmp$StartTime <- NULL
names(tmp)[names(tmp) == "StartTime2"] <- "StartTime"
# ---- Apply the EndTime logic. This is much easier.
tmp$EndTime <- tmp$visitTime2
tmp[tmp$visitTime != tmp$visitTime2,]$EndTime <- tmp[tmp$visitTime != tmp$visitTime2,]$visitTime
tmp$SampleMinutes <- difftime(tmp$EndTime,tmp$StartTime,units="mins")
# ---- We need to be smart here.
# ---- Put the SampleMinutes for the first record for each trapPositionID to -99.
ind <- tmp$trapPositionID != c(99,tmp$trapPositionID[1:(nrow(tmp) - 1)])
if( sum(ind) > 0 ){
tmp$SampleMinutes[ind] <- -99
}
# ---- Put the SampleMinutes for a time frame greater than the gap in fishing length to -88.
ind <- tmp$SampleMinutes > fishingGapMinutes
if( sum(ind) > 0 ){
tmp$SampleMinutes[ind] <- -88
}
tmp$uniqueDate <- NA
# ---- Check where we can. Note that catch.df isn't read in by the function.
# connieSM <- unique(catch.df[,c("oldtrapPositionID","trapVisitID","SampleMinutes")])
# names(connieSM)[names(connieSM) == "oldtrapPositionID"] <- "trapPositionID"
# names(connieSM)[names(connieSM) == "SampleMinutes"] <- "SampleMinutesC"
#
# tmp2 <- merge(tmp,connieSM,by=c("trapPositionID","trapVisitID"),all.x=TRUE)
# tmp2$Diff <- as.numeric(tmp2$SampleMinutes) - tmp2$SampleMinutesC
#
# tmp2[tmp2$Diff != 0 & !is.na(tmp2$Diff),]
# tmp2[tmp2$Diff != -60 & tmp2$Diff != 0 & !is.na(tmp2$Diff),]
# ---- Calculate the proportion of each trapVisitID experience sun or moon, depending.
traps <- unique(tmp$trapPositionID)
sun <- makeSkinnyTimes("sunRise","sunSet",tblDates)
tmp <- getTimeProp(sun,"sunRise","sunSet",traps,tmp,"sun")
moon <- makeSkinnyTimes("moonRise","moonSet",tblDates)
tmp <- getTimeProp(moon,"moonRise","moonSet",traps,tmp,"moon")
# ---- But, we really want proportion of night, and not day.
tmp$nightMinutes <- NA
tmp[!is.na(tmp$sunProp),]$nightMinutes <- as.numeric(tmp[!is.na(tmp$sunProp),]$SampleMinutes) - tmp[!is.na(tmp$sunProp),]$sunMinutes
tmp$nightProp <- 1 - tmp$sunProp
# tmp2 <- F.assign.batch.date(tmp)
#
# tmp.sun <- aggregate(tmp2$sunMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.sun)[names(tmp.sun) == "x"] <- "sunMinutes"
#
# tmp.moon <- aggregate(tmp2$moonMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.moon)[names(tmp.moon) == "x"] <- "moonMinutes"
#
# tmp.night <- aggregate(tmp2$nightMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.night)[names(tmp.night) == "x"] <- "nightMinutes"
#
# tmp.SampleMinutes <- aggregate(tmp2$SampleMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.SampleMinutes)[names(tmp.SampleMinutes) == "x"] <- "SampleMinutes"
#
#
# one <- merge(tmp.sun,tmp.moon,by=c("batchDate","trapPositionID"))
# two <- merge(one,tmp.night,by=c("batchDate","trapPositionID"))
# thr <- merge(two,tmp.SampleMinutes,by=c("batchDate","trapPositionID"))
#
# thr$nightProp <- thr$nightMinutes / thr$SampleMinutes
# thr$moonProp <- thr$moonMinutes / thre$SampleMinutes
#
# test <- thr[thr$trapPositionID == "57001",]
forEffPlots <- tmp[,c("trapVisitID","trapPositionID","StartTime","EndTime","wmForkLength","nForkLength","nightProp","moonProp")]
#forEffPlots <- forEffPlots[order(forEffPlots$trapPositionID,forEffPlots$EndTime),]
#plot(forEffPlots$EndTime,forEffPlots$nightProp,col=c("red","orange","green","blue","black")[as.factor(forEffPlots$trapPositionID)],pch=19)
#plot(forEffPlots$EndTime,forEffPlots$moonProp,col=c("red","orange","green","blue","black")[as.factor(forEffPlots$trapPositionID)],pch=19)
# ---- Rename to preserve, since 'tmp' is used below, and bring in the goodies. Clean up tmp a bit so it merges in nicely,
# ---- and doesn't reproduce data already present in release.visit.
tmpAstro <- tmp
names(tmpAstro)[names(tmpAstro) == "SampleMinutes"] <- "JasonSampleMinutes"
tmpAstro$trapPositionID <- NULL
# drop common columns from merge that happens below
# tmpAstro <- tmpAstro[, -which(names(tmpAstro) %in%
# c( "visitTime",
# "visitTime2",
# "visitTypeID",
# "wmForkLength",
# "nForkLengNth",
# "StartTime",
# "EndTime",
# "JasonSampleMinutes",
# "uniqueDate",
# "sunMinutes",
# "sunProp",
# "moonMinutes",
# "moonProp",
# "nightMinutes",
# "nightProp"))]
release.visit <- merge(release.visit,tmpAstro,by=c("trapVisitID"),all.x=TRUE)
return(list(release.visit=release.visit,forEffPlots=forEffPlots,fl0=fl0))
}
tmcd82070/CAMP_RST documentation built on April 6, 2022, 12:07 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.
Defines functions buildAstroStats
Documented in buildAstroStats
#' @export
#'
#' @title buildAstroStats
#'
#' @description Given a particular trap sequence, calculate the proportion of
#' time spent fishing during nighttime and while the moon was up.
#'
#' @param release.visit A data frame containing release data obtained from the
#' release-data SQL sequence.
#'
#' @param visit.df A data frame containing visits obtained from the catch-data
#' SQL sequence.
#'
#' @return A dataframe with one row for all unique \code{batchDates} included
#' within \code{min.date} and \code{max.date}, inclusive. All variables, as
#' provided via \code{varVec}, have a non-\code{NA} value.
#'
#' @details The \code{buildAstronStats} function first queries the CAMP database
#' for all relevant visits via the \code{Visits} table, as well as the
#' astronomical sun and moon rise and set times, contained within table
#' \code{Dates}. The "Not Fishing" data are also obtained, via a query to
#' table \code{TempSumUnmarkedByTrap_Run_Final} in the underlying CAMP
#' \code{mdb}. Because of this, the catch SQL sequence must be run prior to
#' the same of the efficiency.
#'
#' The function emulates the calculation of \code{SampleMinutes} performed by
#' the catch SQL sequence. This is done in order to calculate denominators
#' for calculation of the proportion of time spent fishing at night and while
#' the moon is up. The \code{SampleMinutes} field from
#' \code{TempSumUnmarkedByTrap_Run_Final} was not utilized because the
#' astronomical metrics need to calculate fishing with respect to the included
#' fishing instances in the calculation of efficiency. These include fishing
#' instances regardless of the value of the \code{includeCatchID} field. The
#' value of \code{SampleMinutes} in the \code{TempSumUnmarkedByTrap_Run_Final}
#' calculated for catch estimation excludes bad fishing, as identified via the
#' \code{includecatchID} field. Therefore, re-estimation of
#' \code{SampleMinutes}, based on efficiency concerns, must be performed.
#' These \code{SampleMinutes} are called \code{JasonSampleMinutes}, so as to
#' differentiate. Values of this variable equals \code{-99} for the first
#' visit of a trapping sequence, and \code{-88} in the case of a time frame
#' greater than the gap in fishing length. Currently, the fishing-gap length
#' is set via global variable \code{fishingGapMinutes}, and equals
#' \code{10080} minutes.
#'
#' Calculation of astronomical proportions utilized \code{campR} functions
#' \code{makeSkinnyTimes} and \code{getTimeProp}.
#'
#' @seealso \code{makeSkinnyTimes}, \code{getTimeProp}
#'
#' @author WEST Inc.
#'
#' @examples
#' \dontrun{
#' df <- buildAstroStats(release.visit=release.visit,
#' visit.df=visit.df)
#' }
buildAstroStats <- function(release.visit,visit.df){
# release.visit <- release.visit
# visit.df <- visit.df
# ----- Define some convenient dates.
min.date2 <- "1990-01-01"
max.date2 <- Sys.Date()
# ---- Grab the time zone.
time.zone <- get( "time.zone", envir=.GlobalEnv )
fishingGapMinutes <- get("fishingGapMinutes",envir=.GlobalEnv)
# ---- Open ODBC channel.
db <- get( "db.file", envir=.GlobalEnv )
ch <- odbcConnectAccess(db)
# max.date.eff <- max(release.visit$ReleaseDate) + 30*24*60*60 # 30-day buffer after
# min.date.eff <- min(release.visit$ReleaseDate) - 30*24*60*60 # 30-day buffer before
max.date.eff <- as.POSIXct(max.date2,format="%Y-%m-%d",tz="UTC") + 30*24*60*60
min.date.eff <- as.POSIXct(min.date2,format="%Y-%m-%d",tz="UTC") - 30*24*60*60
# ---- Get dates information for moon and sun info.
tblDates <- sqlQuery(ch,paste0("SELECT uniqueDate,
nightLength,
moonRise,
moonSet,
sunRise,
sunSet
FROM Dates
WHERE uniqueDate <= #",as.Date(max.date.eff),"#
AND uniqueDate >= #",as.Date(min.date.eff),"#
ORDER BY uniqueDate"))
trapVisits <- sqlQuery(ch,paste0("SELECT trapVisitID,
trapPositionID,
visitTime,
visitTime2,
visitTypeID
FROM trapVisit
WHERE visitTime <= #",as.Date(max.date2),"#
AND visitTime >= #",as.Date(min.date2),"#
AND ( (visitTypeID < 5 AND fishProcessedID <> 2)
OR (visitTypeID = 1 AND fishProcessedID = 2) )
ORDER BY trapPositionID,visitTime"))
# ---- Need to make sure this table exists first, when this code is finalized.
notFishing <- sqlQuery(ch,paste0("SELECT SampleDate,
StartTime,
EndTime,
TrapPositionID AS oldtrapPositionID,
SampleMinutes
FROM TempSumUnmarkedByTrap_Run_Final
WHERE TrapStatus = 'Not Fishing'
ORDER BY TrapPositionID, EndTime"))
close(ch)
# ---- We compile metrics that we need to compile over trapping.
# ---- Construct a trapVisit data frame of all visits.
tmp <- trapVisits
#tmp$fishProcessedID <- NULL
tmp$visitTime <- as.POSIXlt(strftime(trapVisits$visitTime),tz=time.zone)
tmp$visitTime2 <- as.POSIXlt(strftime(trapVisits$visitTime2),tz=time.zone)
# ---- Bring in the mean forkLengths.
fl <- attr(visit.df,"fl")
# ---- Some trapVisitIDs are not in the final catch table, for whatever reason. Find these
# ---- missing trap instances so they can be zeroed out. These trap visits often are found
# ---- in dfs when calculating efficiency. We will put their fish at n of fish at 0.
maxTrapVisitID <- max(trapVisits$trapVisitID)
trapVisitIDSpine <- data.frame(trapVisitID=seq(1,maxTrapVisitID,1))
fl0 <- merge(trapVisitIDSpine,fl,by=c("trapVisitID"),all.x=TRUE)
fl0[is.na(fl0)] <- 0
fl0 <- fl0[fl0$nForkLength == 0,]
tmp <- merge(tmp,fl,by=c("trapVisitID"),all.x=TRUE)
# ---- Construct start and end times. I also construct SampleMinutes...sometimes these differ from Connie's, by 60
# ---- minutes. This has to do with daylight savings. I need POSIX to 'be dumb' with respect to daylight savings,
# ---- to match Connie. I wonder if Connie's SampleMinutes are off by 60 minutes? I suspect the times recorded in
# ---- the CAMP are 'raw' times, and so when one "springs forward" +60 minutes go along for the ride, and when one
# ---- "falls back," CAMP loses 60 minutes. This explains why spring-time SampleMinutes that I calculate are short
# ---- by 60 minutes. I don't think this is easily fixed.
tmp <- tmp[order(tmp$trapPositionID,tmp$visitTime),]
# ---- Apply the lag throughout.
tmp$StartTime <- as.POSIXlt(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone)
# ---- Identify where the lag logic is wrong. These occur when the lag visitTime != lag visitTime2. Adjust the StartTime to be correct.
tmp$StartTime <- ifelse(as.POSIXlt(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone) !=
as.POSIXlt(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime2[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone),
as.POSIXct(strftime(c(as.POSIXlt(NA,tz=time.zone),strftime(tmp$visitTime2[1:(nrow(tmp) - 1)],tz=time.zone))),tz=time.zone),
as.POSIXct(tmp$StartTime))
# ---- So POSIX is awful with ifelse. Or I'm not doing it right. Regardless, put it to how we want it.
tmp$StartTime2 <- as.POSIXlt(tmp$StartTime,format="%Y-%m-%d %H:%M:%S",tz=time.zone,origin="1970-01-01 00:00:00 UTC")
tmp$StartTime <- NULL
names(tmp)[names(tmp) == "StartTime2"] <- "StartTime"
# ---- Apply the EndTime logic. This is much easier.
tmp$EndTime <- tmp$visitTime2
tmp[tmp$visitTime != tmp$visitTime2,]$EndTime <- tmp[tmp$visitTime != tmp$visitTime2,]$visitTime
tmp$SampleMinutes <- difftime(tmp$EndTime,tmp$StartTime,units="mins")
# ---- We need to be smart here.
# ---- Put the SampleMinutes for the first record for each trapPositionID to -99.
ind <- tmp$trapPositionID != c(99,tmp$trapPositionID[1:(nrow(tmp) - 1)])
if( sum(ind) > 0 ){
tmp$SampleMinutes[ind] <- -99
}
# ---- Put the SampleMinutes for a time frame greater than the gap in fishing length to -88.
ind <- tmp$SampleMinutes > fishingGapMinutes
if( sum(ind) > 0 ){
tmp$SampleMinutes[ind] <- -88
}
tmp$uniqueDate <- NA
# ---- Check where we can. Note that catch.df isn't read in by the function.
# connieSM <- unique(catch.df[,c("oldtrapPositionID","trapVisitID","SampleMinutes")])
# names(connieSM)[names(connieSM) == "oldtrapPositionID"] <- "trapPositionID"
# names(connieSM)[names(connieSM) == "SampleMinutes"] <- "SampleMinutesC"
#
# tmp2 <- merge(tmp,connieSM,by=c("trapPositionID","trapVisitID"),all.x=TRUE)
# tmp2$Diff <- as.numeric(tmp2$SampleMinutes) - tmp2$SampleMinutesC
#
# tmp2[tmp2$Diff != 0 & !is.na(tmp2$Diff),]
# tmp2[tmp2$Diff != -60 & tmp2$Diff != 0 & !is.na(tmp2$Diff),]
# ---- Calculate the proportion of each trapVisitID experience sun or moon, depending.
traps <- unique(tmp$trapPositionID)
sun <- makeSkinnyTimes("sunRise","sunSet",tblDates)
tmp <- getTimeProp(sun,"sunRise","sunSet",traps,tmp,"sun")
moon <- makeSkinnyTimes("moonRise","moonSet",tblDates)
tmp <- getTimeProp(moon,"moonRise","moonSet",traps,tmp,"moon")
# ---- But, we really want proportion of night, and not day.
tmp$nightMinutes <- NA
tmp[!is.na(tmp$sunProp),]$nightMinutes <- as.numeric(tmp[!is.na(tmp$sunProp),]$SampleMinutes) - tmp[!is.na(tmp$sunProp),]$sunMinutes
tmp$nightProp <- 1 - tmp$sunProp
# tmp2 <- F.assign.batch.date(tmp)
#
# tmp.sun <- aggregate(tmp2$sunMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.sun)[names(tmp.sun) == "x"] <- "sunMinutes"
#
# tmp.moon <- aggregate(tmp2$moonMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.moon)[names(tmp.moon) == "x"] <- "moonMinutes"
#
# tmp.night <- aggregate(tmp2$nightMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.night)[names(tmp.night) == "x"] <- "nightMinutes"
#
# tmp.SampleMinutes <- aggregate(tmp2$SampleMinutes,list(trapPositionID=tmp2$trapPositionID,batchDate=tmp2$batchDate), function(x) sum(x))
# names(tmp.SampleMinutes)[names(tmp.SampleMinutes) == "x"] <- "SampleMinutes"
#
#
# one <- merge(tmp.sun,tmp.moon,by=c("batchDate","trapPositionID"))
# two <- merge(one,tmp.night,by=c("batchDate","trapPositionID"))
# thr <- merge(two,tmp.SampleMinutes,by=c("batchDate","trapPositionID"))
#
# thr$nightProp <- thr$nightMinutes / thr$SampleMinutes
# thr$moonProp <- thr$moonMinutes / thre$SampleMinutes
#
# test <- thr[thr$trapPositionID == "57001",]
forEffPlots <- tmp[,c("trapVisitID","trapPositionID","StartTime","EndTime","wmForkLength","nForkLength","nightProp","moonProp")]
#forEffPlots <- forEffPlots[order(forEffPlots$trapPositionID,forEffPlots$EndTime),]
#plot(forEffPlots$EndTime,forEffPlots$nightProp,col=c("red","orange","green","blue","black")[as.factor(forEffPlots$trapPositionID)],pch=19)
#plot(forEffPlots$EndTime,forEffPlots$moonProp,col=c("red","orange","green","blue","black")[as.factor(forEffPlots$trapPositionID)],pch=19)
# ---- Rename to preserve, since 'tmp' is used below, and bring in the goodies. Clean up tmp a bit so it merges in nicely,
# ---- and doesn't reproduce data already present in release.visit.
tmpAstro <- tmp
names(tmpAstro)[names(tmpAstro) == "SampleMinutes"] <- "JasonSampleMinutes"
tmpAstro$trapPositionID <- NULL
# drop common columns from merge that happens below
# tmpAstro <- tmpAstro[, -which(names(tmpAstro) %in%
# c( "visitTime",
# "visitTime2",
# "visitTypeID",
# "wmForkLength",
# "nForkLengNth",
# "StartTime",
# "EndTime",
# "JasonSampleMinutes",
# "uniqueDate",
# "sunMinutes",
# "sunProp",
# "moonMinutes",
# "moonProp",
# "nightMinutes",
# "nightProp"))]
release.visit <- merge(release.visit,tmpAstro,by=c("trapVisitID"),all.x=TRUE)
return(list(release.visit=release.visit,forEffPlots=forEffPlots,fl0=fl0))
}
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.